Simple View of Learning Curve

Learning curves are often portrayed in graphs as a single line rising from zero to a higher level with the vertical axis representing some kind of mastery and the horizontal axis somewhat related to time spent learning the material.  The learning of the material is often coincident with the actual development of a product or performance of a job.  So the first parts are usually less well-developed, less sophisticated than later stages.  By the completion of the project or task you are more of a master than when you started.This diagram seems to imply a kind of progress to a success.

Small Failures

Of course some projects do not end in success. These are what are called learning experiences.  People in charge call them successes.  Everyone working for them may consider the project a colossal failure.  People in charge do not like to admit to failure.  Failures are career enders.  Ofttimes people in charge of the people in charge view error as a moral failure.  People in charge leave and go to the next success.  The left behind must use chewing gum, duct-tape and a lot of sweat and blood to keep things going.

Kathryn Schultz author of Being Wrong: Adventures in the Margin of Error.  Very good book.  The gist can be found in the following TED talk:


A Bumpy Learning Curve

Sometimes someone will show a bumpy learning curve.  On the small-scale this indicates a lot of small failures.  Bumping into the misconceptions that plague any endeavor.  The learning curve is displayed as a squiggly rising line.  This is an indication of a trial and error process that goes into real progress on any non-trivial project.   There is a  rise until ignorance is encountered, then a setback, the a rise until yet another area of ignorance is overcome.  This is because for large enough tasks a complete understanding is not there at the very beginning.  The territory is full of unknowns to overcome.

Path From Start to Destination

People try to encode solutions to this in methodologies to avoid the bumps, but cannot avoid the bumps.  The bumps are an indication of a terrain of knowledge that must be navigated during any project until end goal is achieved.  People start at the start and move to a perceived destination.  The terrain is full of unknowns.  Sometimes the project team ends up backtracking over terrain they thought they knew, because they missed something important, that they didn’t realize was important.

The space between start and destination is mostly unknown.  Oops, missed this, oops missed that.  Next time we’ll know better.  Unfortunately next time the terrain is different.

That is why when someone says “We’ll do it right next time” they end up with a bumpy learning curve.  Because the path from start to destination is full of more ignorance and terrain is different.

In the poem about lost soldiers  Brief Thoughts On Maps by Miroslav Holub (a different reference in case the other is gone) the point of the story is often interpreted as ‘any map will do”.  Several respondents on the  monekymagic.net  site brought up the point that the lost soldiers arrived at their own  path to the destination.

Apparently Miroslav Holub credited Albert Szent-Györgyi with the story.

Albert Szent-Györgyi won a nobel prize for discovering Vitamin C and the citric acid cycle. Below is the  first paragraph of the article Lost in the Twentieth Century (found on http://www.annualreviews.org).

It used to be said in my home town, that the cheapest funeral you could get consisted of taking a  candle in your hand and going out, yourself, to the churchyard. As judged by the previous introductory articles, this one should be something like it: an obituary written by the fellow himself. Let it be that. I  have no quarrel with the idea of completing my cycle of life. What goes against the grain is writing about myself, since I  am averse to all forms of exhibitionism. Moreover, I like to look forward, not back.

Alfred Habdank Skarbek Korzybski was fond of saying, “The map is not the territory”.

Maps are the documentation of the terrain.  No one likes to spend a lot of effort in documentation in software development, because the maps are just ideas, concepts, views and other mental ephemera that are constantly  fluxing.  Choosing what to document and how to document things, so that the documenting document isn’t immediately out of date and inaccurate, is a challenge.

Out of words.

There are prettier images of learning curves on from different search engine queries.  Here is a bing query,  a google query, or a duckduckgo query for your confusion.