OnTheSpiral

A couple days ago I was talking to a fellow crossfitter, struggling to explain my daily routine to someone who works a typical 9-5 office schedule. 

I haven't published anything to my blog for a few weeks now. I have been writing (mostly free form) but there has been too much new material in my head that hasn't yet coalesced into anything coherent...too many free floating loose ends. For the past week or two I have been intentionally taking long lunch breaks so as to avoid reaching that point of frustration that tends to sneak up rapidly when I am just getting back into a productivity groove. 

This is what I was trying to clumsily explain when I realized that crossfit (and physical exercise in general) offered a more accessible angle. One thing you quickly learn in crossfit is to never burn out early in a workout...to never go all the way to the point of failure. Let's say I'm doing a workout that involves five rounds of:

Run 400m 
20 push ups
15 pull ups

Push ups are a big weakness for me. If I try to do sets of 20 unbroken I will probably get through the first round without a problem but then reach failure somewhere in the second. Once I reach that failure point I will be burnt out the rest of the way, probably unable to squeeze out more than two or three push-ups at a time for the remainder of the workout. 

If instead I come back from that second run and do a set of ten, shake out my arms for a couple seconds, do a set of five, shake it out, another set of five...I will be able to maintain that pace through all five rounds. There is a body awareness that develops after a while, being able to gauge when you are approaching your failure threshold and stopping just short of that point, beyond which your body will begin to shut down.

That is a near perfect analogy to the management of tempo during challenging cognitive tasks. A reasonable amount of straining builds long term capacity. Too much straining and I cross that failure threshold...frustration sets in, my brain shuts down and short breaks no longer provide a meaningful refresh/recharge. 

*

Beyond this specific analogy, I am intrigued by how frequently I am finding analogies between physical and mental performance. At least for me, physical performance is more transparent/accessible than mental performance and therefore provides a useful lens for studying struggles with procrastination, motivation, energy levels, etc.

Conversely, the analogies almost surely all originate in the brain. People often talk about the brain as a muscle but it would be more accurate to talk about the muscles as an extension of the brain. The physiological phenomena described above have less to do with muscle failure per se and more to do with automatic processes in the brain acting as safeguards against self inflicted injury.

Over the past couple weeks I have moved nearly all of my important/intentional work to paper.  I began to notice nearly a year ago that my writing flowed much more fluidly on paper, as compared to the computer screen.  Since then I have become increasingly aware of all the subtle ways that the screen distracts my thinking.  (As I write this I am working in a tiny G+ text window with a huge youtube embed just below, at least 20 faces [avatars] on screen in various places, dozens of icons of various sorts, etc, etc, etc).

The larger problem is that writing on the screen strips away context by its very nature.  Every word is printed in the same format.  There is no detritus to indicate that I have scratched out and replaced a particular word five times already.  There are no margins in which to leave notes for myself about what I had intended for a particular poorly structured sentence in need of revision.  

There is a finality to writing on screen that derives from the cleanliness of the interface.  When I write on screen I find myself spending more time searching through thesaurus.com than actually composing ideas.  I worry that if I don't express a particular idea correctly the first time, I will fail to notice it during subsequent revisions.  I have tried using various formatting tricks and annotations but such solutions are far more distracting than the equivalent behavior on paper. 

...

Many of the same insights apply to general productivity.  Over the past several years I have experimented with all variety of digital To Do lists, bookmark archives, and numerous other organization/productivity/focus applications.  All have been enormous failures.  The only ones I have continued to use are those that allow me effortlessly capture articles or screenshots (evernote, read it later).  However, particularly in the case of evernote, I rarely ever return to these items in a systematic way.  They basically just allow me to purge things from short term memory that I don't yet know what to do with, without worrying that they are lost forever.
  
Moreover, decontextualized lists in general have almost no motivational impact on me.  I put things on lists, then looks at them later...think to myself, "Yeah I should probably do that stuff"...then I shrug it off and do something else.  The words on the page are too distantly removed from the seat of intrinsic motivation that might actually compel me to take on the task.

I have had dramatically more success as I have moved all this processing over to paper, which has allowed the flexibility to narrate my intentions (To Do's) in full sentences.  The process of narration calls upon both the explicit (what I intend to do) and the implicit (why and how I intend to it).  Tasks only move to a list format after they pass through this self-discussion and I determine that I am actually motivated to complete them.

...  

In the past it bothered me that analog records would be lost...that they would never be transcribed into digital format and therefore would likely be lost, destroyed, or otherwise remain unreferenced in the future.  Recently I have accepted that archiving for future reference is besides the point.

Any worthwhile material that emerges from analog writing will eventually be incorporated into a blog post or other digital artifact.  The final, high quality product will be available for future reference.  The rest of it is just intermediate cognitive processing.  It is the noise produced along the way.  It doesn't need to be saved for future reference.  The noise is not supposed to be neat and clean and searchable.  It simply needs to be externalized so it can be sifted for diamonds.

...

I am curious whether other people have converged on similar conclusions...   

The following is partly inspired by Venkat's recent post - How the World Works: Part II (my comment here).

Innovation cycles oscillate between periods of bottom up innovation and periods of top down innovation.  Favoring one over the other for an extended period of time calls forth the black swan gods...

Take the recent housing crisis...certain "innovations" in mortgage writing, such as variable rates, were initially developed bottom up (setting aside for the moment gov't policies that encouraged lending to low income home buyers).  At the same time, securitized financial instruments like mortgage backed securities emerged to create liquidity for variable rate mortgages.  At small scale there was little danger in this sort of activity.  But as the market developed an appetite for MBS and similar securities, mortgage originators on the ground began to favor them more heavily.  Home buyers who had no business messing around with new fangled mortgage products were pushed into those types of loans because that is what the market wanted sell.  Over the course of several years diversity and resilience in the mortgage market was destroyed...the market became brittle and unstable.  

Admittedly the real estate market involved significant top down intervention from the very beginning so let's consider a less obvious but more pure example.  When smartphones first hit the market, software development for these products was all over the place.  Device makers produce much of the software themselves.  Carriers added their own bloatware.  Some big brands vied for prominent placement.  Some independent developers tried to work with carriers while others delivered applications through the browser.  

No particular distribution modality dominated until Apple introduced the app store.  Since then the majority of mobile development resources have been pulled into the app markets (apple first, now android).  In the short run that evolution of the market was a major innovation.  Buyers of applications and developers of applications knew where to find each other in one centralized place, leading to dramatic growth in the market.  However, in the long run the app stores will become impediments to innovation if they capture so much of the market that other paradigms are ignored or stifled.  

The takeaway from this preamble is that innovation, and economic development in general, must cycle between periods of bottom-up exploration and top-down exploitation.  A focus on only bottom-up innovation will create a chaotic and disorganized market that never fully delivers on its potential and eventually peters out due to lack of coherence.  An emphasis on only top-down innovation will lead to a brittle monoculture that produces a short-term windfall at the expense of future resiliency.  

Debates in economics tend to take on a dialectic character - markets vs central planning, populism vs corporatism, open source vs corporate walled gardens.  These debates present a woefully oversimplified picture of the world.  As human society advances the institutional marketplace broadens and deepens, increasing the range of dialectic conflicts.

Over time, bottom-up institutions tend to grow in capability and influence, taking on ever larger roles.  The most successful of these become top-down institutions themselves by nurturing a new round of bottom up innovation.  Unfortunately, the temptation for the mature top-down institution is to stick its grubby hands in the honey pot to try to exploit the new round of bottom up innovators.  This is where the dialectics emerge.  

Governments spawn markets and corporations but then attempt to milk the market and increase governmental influence.

Corporations create platforms, nurturing communities of customers and commercial partners, but then attempt to lock-in those partners in order to maximize profits.  

In both cases the top-down institution is killing the golden goose.  Their influence should be getting wider but shallower as they move up the stack.  Instead, by trying to deepen their influence they are stifling their progeny.  

Top-down institutions are much like human parents.  They give birth to new entities.  During the early years their children require nurture and guidance.  As the children mature they need to be set free to achieve their potential.  If parents raise their children well they will retain some of their former influence and be taken care of in old age.  If they are overbearing their children will rebel and eventually disown them.

A dynamic yet structured economic ecosystem should behave like an enlightened multi-generation family.  Perpetually renewed by younger generations (emergent institutions), yet receptive to the wisdom of the elder generations.

(Unfortunately the intended video won't embed correctly.  Follow this link to see the video referenced below.)

Above is a somewhat slow-moving but deceptively important presentation. W. Brian Arthur argues that the accumulation of digital processes (processors, sensors, actuators) that can talk to each other are creating a semi-autonomous second economy...an economy that produces value for the human economy with minimal human intervention.

The implication I want to focus on at the moment is the analogy to the industrial revolution and the associated challenge to employment. The argument that automation is a primary cause of unemployment is clearly gathering momentum at the moment. The same argument could have been (and surely was) made 100 years. Industrial/Mechanical technology greatly diminished the need for human labor and threatened to leave vast swaths of workers unemployed.

Yet, in the long run that is not what happened. Short term unemployment gave way to a long term structural shift in the nature of work. Industrial labor was replaced by office work. For the 19th century factory worker predictions of this impending shift would have sounded absurd. People will be paid to sit in comfortable offices and push paper around all day? But of course that was exactly what happened...and as it turned out, it was neither utopian nor dystopian. Office work had its benefits but also obvious shortcomings.

We are in a similar situation today. Clearly digital automation is taking on many of the tasks that are currently major centers of employment. We also seem to be passing through an inflection point wherein change is occurring faster than our ability to structurally adapt. However, in the long run there will surely be some value that human intelligence can add. Given our current perspective, speculation regarding that human value-add might seem absurd.

"People will get paid to do what?!?"

But that is exactly what history predicts. The human economy will do whatever the automated economy cannot. That role might seem trivial today but it will seem critical and obvious in retrospect.

Income distributions are bounded on one end but not the other. You can't earn less than $0, but there is no absolute non-arbitrary limit on the high end of the distribution. As economic freedom increases the distribution can only grow in one direction.  And let's not pretend that there is any meaningful difference between economic freedom and any other form of freedom.  (What use are any other freedoms if individuals cannot direct them towards productive activity and potential commensurate rewards?)  

Moreover, as technology improves the minimum "survivable" income inevitably decreases (inflation aside).  Those same technology improvements are allowing more diverse working arrangments facilitating more granular trade-offs between leisure and labor.  The combined result is that more people will choose to earn less than their potential income and the gap between income maximizers and leisure optimizers will compound (and be conflated with) true/unchosen inequality.