Charlie Alfred’s Weblog

Architect as Guardian Angel and Spinal Cord

The prevailing belief is that the job of an architect is to formulate the architecture of a system.  This architecture is made up of a set of decisions to address challenges which are highly important, difficult, and/or central.

Some take a broader view of the job of an architect.  Ruth Malan wrote:

“Architects make invisible connections and relationships visible — to sense and probe, to identify, to classify, to explore, to navigate, to relate, to understand, to express intent, to describe, to communicate, to test, to ask what if?, to get ideas, to imagine, to reason, to collaborate, to remember, to create and evolve…”[1].

These things are true.  And if you have trouble remembering the long list, here’s a shorter one.  Architects formulate decisions to overcome challenges and, more importantly, they are their projects’ spinal cord and guardian angel.

In my years as a software engineer, I have observed three key things which don’t seem to be related, but which turn out to be linked in important ways:

  1. Systems are complex and this complexity is rooted in interconnections, change and chaos.
  2. The pool of scientific, technical and medical knowledge has overwhelming scope.  No individual is capable of mastering its breadth and depth.
  3. Value is not objective.  Value is perceived subjectively.  These perceptions drive our behavior (what we want, what we think, how we communicate) and usually are motivated by self-interests.

It is the linkage between these three things that dictates why architects must act as spinal cords, and must function as guardian angels.

Complexity is Interconnections, Change and Chaos

Systems are collections of interrelated components with emergent behavior.  Without the help of a tornado or hurricane, all of the parts of an airplane, arrayed in a hangar, cannot fly on their own.  Only when properly connected to the fuselage can the engines generate enough thrust for the wings to create lift.

Few things in our world remain static.  The field of dynamics deal studies how systems change over time.  Some things, like weather and traffic just change faster than other things like rock cliffs or furniture.  Some things, like grass lawns and sand dunes have more subtle changes.

Change creates stress on a system, especially on the connections between components and on the potential for emergent behavior.  Disruptive innovation is a great example of this.  Harvard Professor Clayton Christensen writes:

A disruptive innovation is not a breakthrough improvement.  Instead of sustaining the traditional trajectory of improvement in the original plane of competition, the disruptor brings to market a product or service that is actually not as good as those that the leading companies have been selling in their market. Though they don’t perform as well as the original products or services, disruptive innovations are simpler and more affordable.[2]

As disruptive innovations scale the learning curve, they eventually compete with the established product on cost and value.  Some noteworthy examples include:

  • cell phones disrupting land line phones
  • personal computers disrupting minicomputers and mainframes
  • smart phone cameras disrupting Kodak film.
  • iTunes disrupting brick and mortar music stores

Chaos is a change driving by random variation.  If change brings stress to a system, then chaos amplifies the stress.  Internet hackers and viruses are a form of chaos.  In the NFL, a well-timed blitz brings chaos to offensive lines and quarterbacks, and often the result is a turnover or significant loss of yardage.

Change and chaos make life interesting (perhaps as in the Chinese proverb, “may you live in interesting times”).  Imagine the alternative of pure perfection.  On the positive side: everywhere would have San Diego weather, cars would never get red lights and people wouldn’t get sick.  On the other hand, the population would grow unbounded, tennis points would never end, and all chess games would end in stalemate.   Absence of variation is a precondition of perfection.

The Knowledge Explosion

The size of the pool of available scientific, technical and medical knowledge is staggering.   Researchers at the University of Ottawa Post-Graduate Studies department have estimated that the cumulative total of peer-reviewed journal articles will have reached 50 million in 2009.[3]

The expansion of the knowledge base shows no signs of slowing down.  Mark Mabe estimates that there were about 25,400 active scholarly peer-reviewed journals in 2009, collectively publishing about 1.5 million articles a year.  For the past 200 years, the number of journals has grown by 3.5% per year and the number of articles published by about 3% annually.[4]

These statistics don’t even include other sources of knowledge content such as the vast quantity of professional books, text books and non-peer-reviewed articles.  Nor do they include the increased accessibility offered by personal computers, smart phones and the World Wide Web.

The Gartner Group reported in 2008 that the number of personal computers in the world had reached 1 billion and was expected to reach 2 billion by 2014.[5]  Strategy Analytics projected that the number of smart phones reached 1 billion in 2011, and is expected to top 2 billion by 2015.[6]  With the vast number of connected devices, powerful search engines like Google can locate relevant content, and enable viewing and download in seconds.

Education in the U.S. is challenged to keep pace:

  1. Many graduate high school, but few go beyond.  The 2000 U.S. Census reports[7] that of those 25-44 who graduated high school, less than a third graduated college and less than 10% earned an advanced degree.
Age 25-44 Millions % of Total % of HS
U.S. Population



HS Graduate




Bachelor Degree




Graduate / Professional




Due to knowledge base expansion, a high school education in 1950 was worth more for career prospects than a high school education in 2000.

  1. Education is being compressed, but not fast enough.  In 2001, the College Board administered 3.5 million Advance Placements exams to 2 million students, (over 45% were seniors).  Both numbers are up nearly 250% from 2001.[8]
  2. The knowledge base is too broad and deep to master.  People have little choice but to choose between:
    • Generalist                       conversant in many subjects
    • Specialist                         expert in a few subjects
    • Multi-disciplinarian      capable in several related subjects

In the next two sections, we’ll explore why this makes such a difference for communication.

Beauty is in the Eye of the Beholder

In economic theory, people make decisions based on satisfying (or maximizing) their self-interest.  This is why we look for quality at a low price, buy things on sale, and most importantly, why “choosy mothers choose Jif.”

But value is not as objective as the economists say.  Value is based on subjective impressions and perceptions.  Consider the case of a health 25 year old male who is considering a $500,000 term life policy.  The rates[9] might look something like:

  • 10 year term: $14/month guaranteed
  • 20 year term: $21/month guaranteed
  • 30 year term: $30/month guaranteed

In theory, the cost of the insurance is based on industry-standard mortality tables, with some overhead and profit margin factored in by the provider.  The total cost of the 30 year policy (allowing for the negligible time value of money in 2013) is 30 * 12 * 30 = $1,080.  This puts the odds of a health 25 year old dying during that interval at slightly less than 500 : 1.  The Chicago Cubs, who haven’t won a World Series in 100 years have a better chance of achieving this goal in the next 30 years.

A single 25-year old Chicago Cubs fan, living in his mother’s basement on West Waveland Avenue, is unlikely to put much value in the insurance.  First, he lacks a real beneficiary.  His mother owns the 3 story house they live in (and whose roof he watches the Cubs’ home games from) and is set financially.  He is an only child and doesn’t have a significant other.  Furthermore, the $1,080 premium could be used to erect bleacher seats on the roof of his mother’s house, and charge Chicago tourists for tickets (if only the Cubs could start selling out again).

On the other hand, a 25-year old married man with a wife and two children living in a 2 bedroom apartment in River North has a different perspective.  He is the sole wage earner for his family, and wants a better life for his kids.  $500,000 of term life insurance seems like a good deal.  If he lives to age 55, his kids will both be in their 30’s and he will provide for them.  If not, $500,000 will see to it that they both can go to college.   From an objective point of view, the insurance policy is the same.  But from a subjective point of view, it couldn’t be more different.

Not only is value subjective, but it also tends to be selfish.  Yes, goodwill does exist, and people can be altruistic as well as selfish.  However, altruism tends to be greatest when people feel that their own needs are satisfied.  In addition, much of what passes for altruism can have some underlying selfish motives:

  • A person does a favor for their neighbor, thinking the neighbor might be more inclined to return the favor when they need help.
  • A celebrity donates time and money to a charity thinking that the media attention and publicity can further their career.
  • A sales person donates a large sum of money to a local charity golf tournament, hoping to meet some prospective business contacts.

My goal is not to be a buzz-kill here, and create the impression that everybody is out for #1.  But the point I am trying to make is that people pay attention to and value what they think is best for themselves.  And no place is this more evident than in the way people communicate:

  • One of the number one rules of sales is “listen for what your customer needs and convince them that your product (or service) satisfies their needs better than the customer.
  • One of the number one rules of public speaking and presentations is “understand your audience and tailor your presentation to what they know and what they are interested in.”
  • A person who wants to make a good impression on a first date is given the advice: “get them to talk about themselves and listen carefully

The reason for these similar pieces of advice is that people naturally want to talk about what interests them, they want to talk about it in language that’s familiar to them, and if somebody else starts talking about something that disinterests them, they tune out.  If you doubt, observe a college science or engineering lectures.

There’s a joke that sums this up:

A best man stood up at a wedding to give the toast.  Knowing that the bride and groom had been friends since childhood, the best man said to the 200 people in the ballroom:

“Men, I want you to stand up right now and go stand next to your best friend in this room.”

The bartender was nearly trampled to death.


Now, I’d like to try to connect the topics of system complexity, knowledge explosion and the selfish aspects of value perception.

Most of what we accomplish nowadays, or try to build, is a complex system.  It doesn’t matter if it’s a revised health care system, a reduction in the federal budget deficit, or winning the office NCAA basketball pool.

Virtually all of the simple problems were solved many years ago.  More often than not, today’s problems are “wicked problems.”  These are “difficult or impossible to solve because of incomplete, contradictory, and changing requirements that are often difficult to recognize.”[10]

The knowledge explosion is both cause and amplifier of wicked problems.  It is the cause, as it gives us insights into the deficiencies of existing solutions, and the technologies to do improve.  It is the amplifier, because as Gerry Weinberg points out, “every new system is maintenance for some larger system[11]”, and needs to find a way to fit into the constraints of the systems with which it integrates.

The knowledge explosion interacts with the selfish aspects of value perception.  Figure 1 illustrates how shapes in the game of Tetris help show different types of knowledge appliers:

Testris Example

Figure 1: Using Tetris shapes to illustrate types of knowledge appliers

In Tetris, points are scored by rotating and placing shapes to fill complete lines across the game board.  This placement mirrors real life, as no individual has enough breadth or depth to be effective alone.

But proper placement is not enough.  You can’t just put the marketing generalist in the same office as the computer whiz and expect a killer app to emerge.  No.  Systems possess emergent properties because of interaction and collaboration.  Just as bees spread pollen among flowers, knowledge needs to be transferred and it must show up as comprehension at the destination.

To emphasize this point, consider the analogy between knowledge and electricity.

  1. Stored in one place, electricity is a battery:  potential energy, not kinetic.
  2. Connect a copper wire to the opposite terminals of the battery and you have a circuit.  Electricity flows, but does no work.
  3. Connect one end of a wire to each pole and the other end of each wire to a fan, and the electricity causes the fan’s blade to spin.
  4. Split one of the wires and connect a switch, and you can cause the fan to alternate between behavior #2 and #3.

Electricity is knowledge, and the wires provide communication.  Voltage in the battery causes electricity to flow.  Current gives it the ability to do work.  The fan provides resistance to the current, but not enough to prevent the blade from turning.  The switch enables current to flow through either of the circuits.

The fan provides a productive form of resistance.  Impedance mismatch[12] is a term used to describe an unproductive form of resistance.  Originally defined in electrical engineering, it is the opposition to the flow of energy from a source.

An impedance mismatch occurs if the power source delivers more current to the fan than the fan can handle.  The excess can burn out the wire or the fan.  The opposite is also true.  If the fan provides more resistance than the power source supplies current, the fan won’t turn, even if the circuit is properly connected.

When applied to knowledge transfer, impedance mismatch can obstruct the flow of understanding, due to:

  • the receiver is not paying close (or any) attention
  • the receiver does not understand (or incorrectly understands) the sender,
  • the sender is going too fast and the receiver only gets part of the message
  • the receiver has closely held beliefs and rejects the sender’s message

Real-World Example

In “The Most Expensive Weapon Ever Built”[13] Mark Thompson wrote a scathing critique about the F-35 program.  He describes the story behind a fighter jet program that is $200 billion over budget and already 3 years behind schedule.  The F-35 program is a large, present day example of the topics discussed earlier:

  • A system with large scope and a vast array of interdependencies
  • Changes in context affect the validity of central tenets
  • Massive complexity limit any stakeholder’s view of the whole picture
  • Narrow and subjective perceptions fuel conflicting self-interests
  • Politics and leverage displace communication and comprehension, as the primary driver for decision making

I realize that it may be difficult for you to believe that such things could take place in 2013 in our Nations’ Capitol.  However, read the synopsis below, or better yet, read Mark Thompson’s article, available for free on line.

The F-35 is a 3-in-1 product line design for the U.S. Military and its allies.   This fifth-generation fighter jet program began in 2001 with a budget of around $200 billion, and full production for the last of the three services (Navy) was scheduled for 2012.  In 2013, F-35 program is late, and budget has nearly doubled.  Delays have forced the Air Force and Navy to spend $5 billion to extend the lives of existing aircraft.

Thompson documented several major challenges that the program faced:



One aircraft had to be designed to perform multiple missions

  • Stealth (Air Force)
  • Vertical landing (Marines), and
  • Carrier-based operation (Navy).
A stealthy jet requires sacrifices in range, flying time and weapon-carrying capability.  This poses a severe tradeoff for the Navy, as it forces aircraft carriers much closer to combat Complexity,Tradeoff
The F-35 began production while blueprints were still in flux (10 changes per day). Dynamics,Chaos,


Drones, standoff weapons and GPS-guided bombs have undercut the value of short-flight, piloted, stealth fighters. Dynamics,Chaos
F-35  has 24 million lines of computer code, including 9.5 million on board the plane, over six times the Navy F-18. Complexity,Dynamics

Thompson’s documentation of the program challenges was only outdone by his documentation if the self-interest that amplified these challenges:

  • Justin Amash (R) MI leading the charge for the Congressional deficit-busters.  “Why can’t we spend only 42% of the world’s defense budget instead of 45%”
  • Facing the prospect of sequester, because of Congress’ inability to reach a budget compromise, the Pentagon escalates $5 billion in contractual spending.
  • Each branch of the armed forces could threaten to back out of the F-35 program to gain political leverage (well-known to children as “If I don’t get to bat, I’m going to take my ball and go home.”
  • 48 members of the Joint Strike Fighter Caucus pocketed twice as much as nonmembers in campaign contributions from the F-35’s top contractors during the 2012 election.
  • The constituents of those lawmakers hold a large share of the F-35 program’s 133,000 jobs.

Architect as Spinal Cord and Guardian Angel

The spinal cord carries messages from the brain to muscles, tendons and organs, and carries sensory messages back.  If the spinal cord is injured in the 3rd or 4th vertebrate area, the result is often paralysis from the neck down.  Autonomous operation of the heart and lungs is all that keeps this injury from being fatal.

An architect is called on to make difficult decisions.  These decisions address the biggest challenges facing their project: those that deliver important value to stakeholders, address difficult tradeoffs, overcome major risks (like security and safety) and make the system resilient to contextual variation and change.

As systems become larger and more complex, they demand high performance, fault tolerance, concurrency and integration with other systems.  This requires deep subject matter knowledge of many domains.  Even the most competent architect will have difficulty mastering them all.

An architect’s job usually required collaboration with stakeholders whose knowledge is broad and shallow and specialists whose knowledge is deep but narrow.  This is where the spinal cord aspect of the architect’s role becomes critical.  It requires effective transfer of knowledge in both directions, even in the face of blank stares, resulting both from disinterest and lack of understanding.

To be clear, effective communication is a two-way street.  Stakeholders and specialists are challenged to engage fully, and communicate beyond their narrow areas of interest.  However, it is the architect’s responsibility to enable this level of engagement, by:

  • Listening carefully, for messages beyond the words used.
  • Probing for why things are said, rather than settling for what or how
  • Explaining concepts in terms the other can understand:
    • translating technical jargon for non-technical stakeholders, or
    • explaining the impact of obstacles or risks from other domains, and relating them directly to the concerns of technical specialists.
    • Asking probing questions to understand priorities, dependencies, variations and justifications.
    • And most importantly, ruthlessly seeking and flushing out the “unknown unknowns.”  Blind spots are deadly.

The spinal cord role is different from but closely linked to the guardian angel role.  In spite of the architect’s huge responsibility, authority to make critical decisions often rests with others, including: executives, product managers, customers, and sometimes the board of directors.

An architect is required to be a persuasive lobbyist and influencer.  Again, this demands powerful communication skills – in expression, but especially listening and reading.  The most important clues for how to make a point are usually found in actively recognizing them.

The other guardian angel role comes from being a lookout for the project.  Most critical changes, threats and originate outside of the project.  Dealing effectively with these requires three skills:

  • recognizing their occurrence, and
  • understanding the (potential) significance of their impact, and
  • being able to communicate this significance to others whose attention is elsewhere.

The lookout on the Titanic recognized the icebergs, understood the threat they posed, but could not convince the captain to change course until it was too late.


Systems are “wicked problems.”  They have many powerful inter-dependencies, constraints, tradeoffs and risks.  They change frequently, often in chaotic ways.  The pool of knowledge in our world is too broad and deep for any individual.  Success requires effective collaboration between generalists, specialists and multi-disciplinarians.  Unfortunately, people are self-centered and value what they perceive to be important.  Often, this is limited by what they already know and believe.

Architects have the difficult job of acting like the project’s spinal cord.  Collaboration requires transferring messages between people with different levels of understanding and disjoint mental models.  More important than relaying the message is figuring out a way to ensure that it is understood.

If this isn’t enough, architects also have the job of being guardian angel.  They must have patience and perseverance to ensure that good decisions are made, even when they lack the authority to make them themselves.  In addition, they need to be on the lookout for external factors that might threaten the system, and communicate with stakeholders and specialists to ensure that the impact posed by these factors is understood.

Wonderful Life

Henry Travers as Clarence (the Guardian Angel) and James Stewart as George Bailey (the stakeholder) in It’s a Wonderful Life


[2] C. Christensen; J. Grossman M., ; J. Hwang M.D.,  (2008). The Innovator’s Prescription : A Disruptive Solution for Health Care. McGraw-Hill. Chapter 1.

[3] Arif Jinha  “Article 50 Million: An Estimate of the Number of Scholarly Articles in Existence”, University of Ottawa

[4] Mark Mabe, “An overview of scientific and scholarly journal publishing”, 2009 International Association of Scientific, Technical and Medical Publishers

[7] Source: U.S. Census Bureau.

[11] G. Weinberg, Secrets of Consulting, (1986) Dorset House

[13] M. Thompson, “The Most Expensive Weapon Ever Built”, Time Magazine, February 25, 2013,,9171,2136312-6,00.html

1 Comment

  1. […] I believe that this is the next frontier in systems development, and that this will be a major part of the architect’s job (if if isn’t already).  To read more about why I believe this, check out: […]

    Pingback by Communicating Understanding – the Architect’s Challenge | Charlie Alfred's Weblog — March 10, 2013 @ 12:25 pm

RSS feed for comments on this post. TrackBack URI

Blog at

%d bloggers like this: