All We Like Sheep Have Gone Astray

PRINCE2

Unsettled, the haplessly-institutionalised incumbents huddled together for warmth, two to a cubicle, cruel flickering artificial light obliterating any capacity they once might have harbored for self-actualisation or for problem-solving, casting an especially-stark relief upon their miserable, wasteful, angsty souls.

They were unsettled by the juggernaut paperweight reputation of the new shepherd coming to guide them, probably to slaughter them, certainly to restructure them, and this new shepherd, this new shepherd was certified!  This new shepherd with a string of certifications long as a donkey’s tail: TOGAF, ITIL, PRINCE2, PMI, COBIT, MSCE, and a beginner’s first-aid certificate and some first-year classics and ancient history university study as a bonus.

Their fearfulness was palpable and paralysing and visceral, but, in the end, thoroughly unwarranted, for this new shepherd was just a person, just a person with the same pains, hopes, and fallibilities the rest of us have.

There continues to be much over-hype about certifications in IT, particularly about enterprise-architecture certifications, and particularly about TOGAF certification.  Demonstrating core knowledge in a domain or discipline through a standardised assessment process is a good thing, and establishing reliable credentials is a good thing for the profession of Enterprise Architecture.

Where the simple system of “credentialled = worthy” breaks apart is the point at which the credential itself becomes imbued with mystique and desirability and is viewed as an end in itself.  For enterprise architecture, where the professional practice must be adapted from one vertical to the next, from one organisation to the next, from one practitioner to the next, and from one type of engagement to another, having a detailed map to follow is neither possible nor sensible.  What’s needed is a compass, not a map1, and none of the major enterprise-architecture frameworks currently provide compasses or maps.

Gartner summarises the current state of affairs nicely in the 2014 edition of the Hype Cycle for Enterprise Architecture2, including observations that:

  • certification’s value is unclear to organisations
  • certification’s value is unclear to individuals
  • certification is a good thing to have on your resume
  • certification implies almost nothing about proficiency

The always-insightful, always-thought-provoking Tom Graves makes crucial observations about skills and (or versus) certifications in his Saving enterprise-architecture from itself post3:

  • certification has become a commonplace recruitment filter for enterprise architecture roles, and this approach weights certification (which is typically no more than a few-days-long short course) overwhelmingly and disproportionately more heavily than years of applied experience working as an enterprise architect
  • certification in enterprise architecture frameworks ignores the enormously-important complement of soft skills (the communication, leadership, collaboration, and adaptability) that are essential for success as an enterprise architect — they do this because the enterprise architecture frameworks themselves are barren of soft-skills content

Jason Bloomberg’s new Bloomberg Agile Architecture4 is a great new example of practical, adaptable, and flexible applied enterprise architecture.  This new technique differentiates itself from the traditional heavyweight do-everything subtractive frameworks (i.e., you discard pieces of the framework you think  you don’t need) by instead taking an additive approach that incorporates elements, artefacts, and capabilities as they are found to be required in order to deliver a particular business outcome.  There’s much to like about this approach.

The danger for enterprise architecture is that the academic energies wasted mewling sheeplike after the frameworks and the certifications are preventing us from forming constructive connections with our customers, partners, and stakeholders, and inject bedazzling interfering distractions that prevent fulfilling the mission and promise of business-outcome-focused enterprise architecture.  Curiously, the more of us that become certified in a common framework, the better the chances of those frameworks being improved, of architects developing common vocabulary and process, and the better the chances of certification’s value becoming clear and visible.

The frameworks have much to contribute to and provide anchor for the practice of enterprise architecture, and there is value to be found in them, but leading in a still-diverse and still-immature profession with overemphasised focus on the merits of certification as a rite of passage and entitlement to speak is leading all we, like sheep, astray.

 

 

 

 

  1. Principle 3 here is very nice: “compasses over maps” = MIT-Knight Civic Media Conference, 2014, Joi Ito’s 9 Principles of the Media Lab, http://vimeo.com/99160925
  2. Betsy Burton & Philip Allega, 2014, Hype Cycle for Enterprise Architecture, 2014, http://www.gartner.com/document/2804819
  3. Tom Graves, 2013, Saving enterprise-architecture from itself – 3: Skills and certification, http://weblog.tetradian.com/2013/11/11/saving-ea-from-itself-3-skills/
  4. Jason Bloomberg provides essential reading for enterprise architects in these two pieces:
    1. 2014, Is Enterprise Architecture Completely Broken?, http://www.forbes.com/sites/jasonbloomberg/2014/07/11/is-enterprise-architecture-completely-broken/
    2. 2014, Is the BAA Technique an Architecture Framework?, http://intellyx.com/introducing-the-bloomberg-agile-architecture-technique/is-the-baa-technique-an-architecture-framework/

“Fairy Bright Eyes”, Projects, and Decay

Image

“Fairy Bright Eyes” is an installation by Ryan Monro, one of a dozen artworks in Auckland’s Learning Precinct Micro Sites initiative1.  A unexpected curiousity in an unusual location on a university campus, this piece represents beautiful aspirations and their subsequent inevitable decay:

  • A chandelier hangs over an alleyway. At first it symbolises luxury and hope but as it degrades over time it becomes a symbol of dystopia.

i’ve recently encountered a number of situations in which problems have appeared or changes are needed to the deliverables of projects that were closed as success stories, and the organisation’s ability to address those problems or make those changes has been wavering between absent and resentful, and for extraordinary, though predictable, reasons including:

  • the project didn’t deliver that function, and we’re forbidden to work on it now as a business-as-usual activity.
  • the project was outsourced, and nobody here knows how it works, and the documentation we were promised never materialised.
  • the project went live, but there was a lot that got moved into Phase 2, but they spent all the money on Phase 2 fixing problems with Phase 1, and now we don’t have anything that was promised in Phase 2.
  • the project delivered a product, but the only person that knows how it works has moved onto another project, and we’re not allowed to speak with him.
  • the project couldn’t deliver that integration because we failed to overcome political problems with the other system owner, and now there is no funding left anyway.

At once, i was reminded of the chandelier and its spectacular decay from a thing of sparkling beauty into an exhaust-tarnished relic, and reminded of two essential takes on the problem with projects and their definition of success:

  • energy, funding, and resource are assembled and brought to bear in the greenfields stage of a project, and those elements, along with the peculiar nature of the early greenfields condition, enable projects to move quickly and make initial changes and delivery with reasonable agility… but subsequent changes are soon required on top of initial changes, and then the project ends, and a prolonged stagnation period ensues2.
  • the success of an application project can be judged only by its long-term ability to deliver business benefit and by the change-friendliness of its capability delivery3, which is determined by the extent to which its non-functional characteristics have been understood and quality-assured4.

As an industry, we’re doing a terrible job of delivering application projects.  A big part of the solution to turning this terribleness around and improving these projects must come from the enterprise architecture discipline, embedding good behavoiurs, patterns, principles throughout the project lifecycle, from inception through design and delivery to long-term benefits realisation and, eventually, into managed decay.

 

 

 

  1. Various Artists. (2010) Learning Quarter Microsites, described at http://www.aucklandcouncil.govt.nz/EN/newseventsculture/Arts/publicart/Pages/learningquartermicrosites.aspx
  2. Krafzig, D., Banke, K., & Slama, D. (2005) Enterprise SOA: Service-Oriented Architecture Best Practices. Prentice Hall. ISBN 0-13-146575-9
  3. Enterprise Architecture in 140 Characters, Brenda Michelson, Elemental Links, http://www.elementallinks.com/2010/02/17/enterprise-architecture-in-140-characters/
  4. Kyte, A. (2011) In Application Projects, ‘Success’ Needs Many Definitions. Gartner Research, Article ID #G00210218.

“What value will a Solution Architect add to this?”

tessellations as repeatable solution patternsAfter some prompting, the technical requirements arrived for architectural review.  They had been prepared from a well-intentioned place and under the pressure of project timelines.  Scanning through them revealed serious problems: these requirements were much too far progressed, and were based upon several fundamental anti-patterns.  Needing to deliver some new information from one ERP system to another, the proposed integration solution:

  • specified silent discards in favour of raising alerts when errors were detected
  • favoured unsustainable side-door integration techniques over the use of established patterns
  • was disconnected from the business requirements it has to deliver
  • specifies data-exchange and transformation rules at the wrong layer of the stack

These are unacceptably-bad things, and while their origin is understandable and of an environmental and historical nature, rather than of an individual or other nature, they must be addressed, and the course proposed redirected towards simple compliance with basic established delivery patterns.  This work is not supporting an isolated endeavour, a proof-of-concept experiment, or an early innovation — this is a mainstream production deliverable where there is genuinely no latitude for managed diversity.

Perhaps inevitably, the project machine responded as follows, heavily paraphrased:

  1. why are you making us have to do this properly?
  2. what is the value of involving a solutions architect?
  3. don’t you know we’re busy?

The spectrum of governance ranges from assistive governance through to directive goverance, in both proscriptive (you must not do these things) and prescriptive (you must do these things) flavours.

While there might not always be uniquely-right or best-ever answers and solutions, there is a sense of universal truth and rightness here, and sometimes the architecture function must adopt a directive position, generally when assistive positions have failed to achieve a sustainable business-value-driven outcome.

Facilitating the use of established repeatable patterns, ensuring change-friendly sustainable outcomes, and sometimes veering into directive governance, that’s the value of a solutions architect.

busoto is collaboration

build something together

In 2011, and again in 2012, while preparing to cofacilitate with Ric Phillips1 a workshop about collaboration across an enterprise-architecture community of practice, we were reminded with jarringly clarity that the word collaboration is misunderstood and misused, and that unhelpful behaviours result from this misunderstanding and misuse.  Under the banner of collaboration are in fact three broad classes of behaviour:

  1. Connecting
  2. Sharing
  3. Collaborating

…and the expectations upon participants and on the elements needed to support those behaviours vary greatly from one broad class to the next, as:

  1. Connecting is simply establishing and in some way monitoring connections between people in environments such as LinkedIn and, to a certain extent, Twitter and Yammer.  Connecting gives rise to ambient intimacy2 — knowing what is happening, knowing something of what your peers are doing.
  2. Sharing is making available to others some or all information, decisions, and other artefacts (e.g., strategies, templates, documents, diagrams, code) that are generated by doing work. Sharing can be mediated by websites, blogs, Twitter, and so forth.
  3. Collaborating is the deliberate cooperative creation of something new: new products, new artefacts, new understanding — in any form, new value.

Selected good definitions of collaboration include:

  • two or more people working together towards shared goals3
  • creative coordinated goal-oriented4
  • working together to achieve a goal5

…and three random examples are offered below:

  • busoto is a new portmanteau of the phrase build something together, illustrated by the drawing at the head of this post, which my five-year-old daughter and i built together deliberately.
  • Wai Notes6, 7 is a music album containing early/demonstration versions of songs that were recorded by Will Oldham and Dawn McCarthy by exchanging tapes in the postal mail.
  • A group of architects from different organisations obtain local sponsorship to work together to define and create business capability models for their sector.

Collaboration, busoto, is a deliberate and identifiable behavioral style.  It builds upon, depends upon, and is a higher form of connecting and sharing. Communicators and collaborators must acknowledge and stay actively aware of the difference between connecting, sharing, and collaboration.

  1. Ric Phillips and his excellent eapatterns blog at http://eapatterns.tumblr.com/
  2. Ambient Intimacy coined and defined at http://www.reboot.dk/page/1236/en
  3. Ephraim Freed’s August 2012 post What Collaboration Really Means at http://www.thoughtfarmer.com/blog/2012/08/14/what-collaboration-really-means/ includes a useful list of behaviours often confused with or labelled as being collaborative, but that are not collaboration.
  4. Jane McConnell’s May 2012 post Clarifying the word “collaboration” to reduce confusion and conflict at http://netjmc.com/social-collaboration/clarify-to-reduce-confusion-and-conflict includes some helpful models indicating how collaboration fits into the digital workplace and other work and organisational patterns, alongside intranet/content and people-connectedness.
  5. The wikipedia definition at http://en.wikipedia.org/wiki/Collaboration
  6. Wai Notes at Drag City http://www.dragcity.com/products/wai-notes
  7. Wai Notes at Wikipedia http://en.wikipedia.org/wiki/Wai_Notes

The Architecture For This Project Is Out Of Scope

not my monkey

It is fair to say that the project in question has not started well. A lengthy incubation period executed in isolation from business and delivery partners has seen resources consumed in exchange for very little demonstrable progress, and high-level commitment waning. Faced with challenges like these, project managers will often retreat into their defensive belief systems — the heavily-dogmatic linear methodologies. Under such circumstances, the project manager looked me in the eye and uttered the incredible words:

  • “The architecture for this project is out of scope for this project.”

This statement dumbfounded me, and made me think. Another literal expression had been made of the natural tensions between enterprise architects (and their broad-context thinking and striving for change-friendly capability delivery1) and project managers (and their project-constrained and limited-context viewpoint and striving to achieve traditionally-measured project success). Variously, while i am:

  • accepting that enterprise architecture does not have a strong track record in the successful application of its own heavyweight frameworks
  • holding strong personal convictions that well-executed P3M (Portfolio, Programme, and Project Management2) has transformative potential for good
  • noting signs that enterprise architecture and P3M practices can deliver better business outcomes working together harmoniously3

…i’m too often seeing struggling projects “remedied” through the application of portfolio-grade heavy management and governance techniques, with the primary effects being:

  • greater focus on time and budget, at the sacrifice of scope and benefit
  • more expensive (in terms of both effort, money, and time) project contols
  • blossoming of the infamous opportunities register

For this to change, the culture of projects, and the culture of enterprise architecture, needs to change, meld, and align — a new conversation is needed, needed to lift every project’s context for decision-making up to at least the portfolio level.

It was much later before i established the link with this item doing the rounds on Twitter:

  • “Not my problem” in Polish is “nie moj cyrk, nie moje malpy.” Literally “not my circus, not my monkey.”4

There it is, the crux of the execution-time tension: “not my circus, not my monkey“.

As we step down through the layers of P3M we find portfolios (selected circuses and selected monkeys), programmes (small numbers of circuses, small numbers of monkeys), and projects (one circus and some of its monkeys).

With its enduring concern for the broader context, enterprise architecture is concerned with the enterprise: with all the circuses, and with all the monkeys.

  1. Enterprise Architecture in 140 Characters, Brenda Michelson, Elemental Links, http://www.elementallinks.com/2010/02/17/enterprise-architecture-in-140-characters/
  2. The Project, Programme and Portfolio Management InfoKit from JISC http://www.jiscinfonet.ac.uk/p3m provides a good overview of P3M.
  3. Bittler, R.S. (2012) Best Practices for EA and PPM Integration Toward Improved Business Outcomes, Gartner Research, Article ID #G00237525, http://my.gartner.com/portal/server.pt?gr=dd&ref=shareSummary&resId=2173516 gives a pragmatic and actionable set of best-practice suggestions for aligning enterprise architecture and P3M teams and initiatives.
  4. @howardtaylor on Twitter at https://twitter.com/howardtayler/statuses/260944798167482368

Ideal Circumstances for EA Frameworks

Ideal Circumstances for EA Frameworks

Significant touchstones, at least for some people, some of the time, the series of hype cycle devices published annually by Gartner inform and guide industry behaviours, lead and shape the formulation of strategy.

Three features of the current enterprise architecture hype cycle1 are noteworthy, or thought-provoking, or self-evident:

  1. Whole-of-Government enterprise architecture has been cancelled due to lack of interest and because it is unlikely to deliver transformational benefit… except under ideal circumstances.
  2. Enterprise Architecture Frameworks are on notice, because they’re implemented as a whole insufficiently often to make solid ground, they’re probably a decade away from delivering benefit, and too many enterprise architecture practitioners are distracted by them.
  3. Enterprise Architecture as a discipline “…proactively and holistically leading enterprise responses to disruptive forces by identifying and analysing the execution of change toward desired business vision and outcomes” is moving into the plateau of productivity, and the discipline of enterprise architecture will deliver transformational benefit within the next few years.

Tying these features together and thinking about what they mean for my own enterprise architecture practice generates the following simple good-vs-less-good system:

  • Old School: heavy-framework approaches tend to be neverending documentation exercises capable of producing enormous amounts of exactingly-beautiful and perfectly-irrelevant families of well-organised zero-benefit enterprise architecture artefacts.
  • Change-Friendly: the ultimate outcome of enterprise architecture is change-friendly capability delivery3

…and some unanswered questions:

  • Ideal Conditions: what are the ideal conditions under which enterprise architecture (as a discipline) and enterprise architecture frameworks (as a belief system) and whole-of-entity enterprise architecture (where “entity” might be a division, an enterprise, a government, or a global corporation) may thrive and deliver the transformational benefit they promise?
  • Measuring Benefit: particularly vexed and repeated plaintiffly everywhere, the question of how we as a community of enterprise architecture professionals measure (and communicate well) the benefit delivered by the enterprise architecture discipline.

Whether emergent, pace-layered, pragmatic, lean, agile, or just-enough — by any name, involving enterprise architecture in change initiatives, in strategy formation, in policy review, in service design, in everything, usually leads to improved outcomes.  This happens through a focus on the discipline, values, and core principles of enterprise architecture, rather than through the naked and contextless application of the heavy frameworks.

With the enterprise architecture discipline thriving, and in some areas undergoing a resurgence, we might yet as a community of practice be able to create for ourselves those very ideal conditions under which the frameworks, too, might thrive… what are the ideal conditions for enterprise architecture to succeed?

  1. Burton, B. & Allega, P. (2012) Hype Cycle for Enterprise Architecture, 2012, Gartner Research, Article ID #G00234608
  2. Gall, N. (2012) Gartner’s 2011 Global Enterprise Architecture Survey: EA Frameworks Are Still Homemade and Hybrid, Gartner Research, Article ID #G00226400
  3. Enterprise Architecture in 140 Characters, Brenda Michelson, Elemental Links, http://www.elementallinks.com/2010/02/17/enterprise-architecture-in-140-characters/

Heal the Eternal Disconnect Between Project Managers and Enterprise Architects!

flexible fire loopTypical exchanges between project managers and enterprise architects have historically, anecdotally, and sterotypically been predictable, unconstructive, and unsatisfactory affairs that juggle and irritate tensions across dimensions such as:

  • enterprise context vs project scope
  • change-friendly capability delivery2 vs project delivery
  • methodological and framework differences
  • open/collaborative vs closed/secret

The nature of these exchanges has led to the creation of an eternal-looking disconnect between project managers and enterprise architects.  Flummoxed recently by a representative exchange that ran like this:

  • EA: “those requirements still need a lot of work, not least because they don’t provide the necessary coverage, and because there are mistakes throughout them”
  • PM: “no, we don’t have time to make them better, because we’ll then be behind schedule”
  • EA: “…but they’re wrong, inadequate, and embarrassing, and intended for publication externally to vendors!”
  • PM: “they were circulated to a reference group before you saw them, so, under the project charter, we’re not allowed to change them now anyway”
  • EA: “…but they’re wrong, inadequate, and embarrassing”
  • PM: “it is what it is, so just eat it! you enterprise architects are all the same, you’re all annoying, and not one of you understands what a successful project is, or how important following the correct methodology and controlling scope is to ensuring success”

…i have become certain the fundamental issue here is context.

In terms of the Zachman Framework3, the audience perspectives illustrate the context neatly, ranging from high-level whole-enterprise-and-beyond context:

  • executive = business context planners

…down to constrained appropriately-defined task-scoped context:

  • implementers = business component implementers

Traditional concerns of the project manager are all about landing something, landing anything approved by a steering group, landing an awkward three-legged contraption on the fluid trampoline of a go-live date. Traditional concerns of the enterprise architect are all about understanding how any change is incorporated into the functioning enterprise, and how to ensure that change is sustainable, able to be integrated within the business and technology bedrock of the enterprise, and consists of repeatable patterns and reusable artefacts for good, generating business value beyond the immediate at-hand implementation scenario.

The three legs of the awkward contraption are, essentially: effort, money, and time… but, just as Todd Biske outlined recently and eloquently1, the project-delivery obsession with being on-time and on-budget readily sacrifices scope, scope, the reason for and the justification of the project being commissioned in the first place.

The disconnect spans a greater distance than the relationship between project managers and enterprise architects.  The disconnect is cultural, and relies upon much better project execution, much better portfolio and programme management, and much better procurement processes.  It’s a core responsibility of the enterprise architecture profession to ensure these activities work together well, create business value, and facilitate change-friendly capability delivery.

Ultimately, the disconnect can be healed only by participants with broad-context viewpoints and strong commitment to ensure that change delivers sustainable business value. In this ecosystem, the only suitable relationship leader is enterprise architecture, enterprise architecture doing its job as advertised, and doing it well.

 

 

 

Notes:

  1. Think Enterprise First, Todd Biske: http://www.biske.com/blog/?p=878
  2. Enterprise Architecture in 140 Characters, Brenda Michelson, Elemental Links, http://www.elementallinks.com/2010/02/17/enterprise-architecture-in-140-characters/
  3. About The Zachman Framework at Zachman International: http://www.zachman.com/about-the-zachman-framework