Sensemakers versus Skynet

Esteban Kolsky makes me think.  Like today, he posted a provoking article on Sensemakers versus systems – and he was stimulated, in turn by Venessa Miemis, another favorite thinker of mine.  Let’s hear it for people who make us think.

I started to post a response to Esteban’s article but as I watched the length of my comment grow, decided to post it here instead.

Esteban postulates:

[D]o we need Sensemakers, people who can make sense of the data — or can we trust the systems to make sense and make the decisions for us?

Esteban’s essential argument is the world is too complex and the data volume so high that computers are needed to handle such high volumes of information and then act on them.

My thought?  I agree in part with the Esteban...but with a big qualification.

Today’s complexity

He’s right. Today's world is complex and getting our hands around the information needed to manage it is a challenge.  And the challenge will continue to grow.

What I want from computers is the ability to capture data and the tools that allow me to help analyze it. Pattern recognition is important and often times hard to see within the raw data so the tools need to have the ability to help clear the data clutter.

Additionally, I want to be able to marry information from disparate sources to see additional patterns.

Avoiding Skynet

At the end of the day, though, decisions need to be left in human hands. Not interested in a future Skynet (or the Colossus of my youth).

Jokes aside, Skynet and Colossus point to very real concerns. Computers can't match the subtlety and ability to juggle complexity that the human brain is capable...although human brains need training too in order to manage complexity.

Developing Sensemakers

Sensemakers are needed in this world - and the skill is a combination of innate strengthened with practice. We need to teach our children how to think things through and not learn by rote.

The most important class I took in high school was Algebra because it taught me now to solve equations with unknowns. I still use the skill to this day, even if I don't approach solving problems with a math formula per se.

The most important classes I took in college were my geology classes for my major. Geology taught me how to understand complex systems. Geologic interpretation requires one to look at an area that's been impacted by many events over the course of millions of years. Each of these events pile on top of each other, morphing and clouding the results of previous events. The geologists challenge is to peel back the layers to figure out what happened when.

Some people are naturally talented at complex thought and solving unknowns in the face of uncertainty. And some of this can be taught. And yet more is strengthened with practice.

Sensemakers and collaborators

Last thought...Our world is complex enough now that no one person can think through everything in the "big picture."  In larger challenges, there are too many areas of expertise required for one person to know everything.

That’s why we need to teach our children the art and enjoyment of collaboration at the same time we teach them competition.  I’m not vilifying competition.  It has it’s place in motivating us to action.  But in the wake of the Winter Olympics, I can’t help but think of Lindsey Vonn and Maria Riesch.  Two fierce competitors who are also supportive friends.

My 2 cents.

Comments

[Esteban Kolsky]

Kathy,

Very interesting piece, i think it truly takes the conversation forward. I am still trying to respond to the comments in my blog in regards to this matter, but I think that the biggest problem people have is that by losing the human in the equation we lose the qualities that makes us not-machines.

Unfortunately, the massive amount of data we are accumulating must be processed both in a timely manner and with relevant results highlighted. And there are no humans that can do this, only systems (the key of this sentence is timely processing, not massive amounts).

The concept of sensemakers that Venessa brings to the table is something that I heard before and read lots about both the in-between link from massive processing to human filter to analytical process, or from massive and analytical process to human filter. I even wrote some of these ideas into a model I used to call secret customer service where massive processing met the human filter before committing an answer to another human (without letting the recipient know that it was "human approved"). It has been tried and never gotten too far because the scale and knowledge workers are not there to make it work.

thanks for furthering the discussion, and for partly agreeing with me. I think that time will prove me right -- just not sure what timeframe it is :)

Esteban

[Kathy Herrmann]

Esteban,

You make a good point. Having apps that can help accelerate the interpretation of data is a good thing. Pattern recognition software and such goes a long way towards that goal.