Learning about Learning

I’ve spent a lot of my recent career solving problems, and thinking about how to go about solving problems – methodology for solving problems.  I have developed a leaning towards lean / agile problem solving which encourages fast experimentation, intellectual honesty (understanding that you don’t know the answer and that you’ll probably get it wrong the first time around) and reflection.  The faster you can perform this loop the faster you learn.

Improving the methodology is all about figuring out how to run experiments, get results, learn : and doing all of that as fast as possible and at scale.

And so I found myself applying this way of thinking to bonsaiponics and i’ve figured that i’m going about this all wrong, or at least I have now reached the point where I need to change the way i’m experimenting.

Up until this point all my learning has been about picking up initial subject matter expertise; does this grow, does this work, what will fish eat, what temperature water, how to hook up system, how to sow plants, how to filter water, nitrogen cycle.

All of this is good solid initial subject matter expertise.  However, if I want to get beyond small experiments I need to start figuring out how to do all of this at scale.  The challenge with nature is that experimentation takes time, plants take time to grow, systems take time to balance, learning in nature in an asynchronous fashion will take a long time. Plus there’s the seasons.  So, i’ve got a window to get a lot of learning done. How can  achieve this as effectively as possible.

The thought process:

So this is how I figured what  had to do….

I reasoned with myself, and recognised that i’m not going to get it all right, there’s so many levels of complication that I still don’t know about, that although i’ve removed several levels of uncertainty…there are still many layers of the onion left.

If I continue in this way i’ll have a broad (horizontal) but shallow knowledge over several things, but not much practical results to show for it.  However, if I can focus, go deep on a limited scope, more vertical than horizontal, I might get some practical results and eventually success.

This reminded me that I had not yet outlined what the success criteria were – what did success look like for me?  In IT talk that would be the V in MVP (Minimal Viable Product) – What is the minimum thing that could happen that would make me feel like this was a success?

What is Success?

For me it was all about production and creation.  At the end of the day, I wanted to produce some food.  I wanted to produce food at a meaningful scale that it could replace my need to buy certain vegetables.

In the past, i’ve planted and had some success with several different types of vegetables, but I never produced enough to make a meaningful indent in to my actual habits. This was the key.  If I could determine what would make me feel successful at the end of the year, then I could use this goal to reduce the scope of learning needed and focus.

The Target:

So, I decided that I was going to focus on 3 core crops which were stables in our house: Salad, Tomatoes, and Radishes.  I then defined the scope of success to mean that I would want to be producing at least, 1 salad, 6 tomatoes (equivalent mid sized), and 20 radishes every week for as along as I could.

There was still a little uncertainty (subject matter expertise) about how long it would take me to get up and running, and exactly how long the crops would produce for, so I did not define how long I had to maintain this target.  However, as a guide I set a duration of “at least one month”:  meaning that at some stage of the year i’d be producing to my target sustainably for 1 whole month (4 consecutive weeks).

Right with that decision made, I immediately started to do some back of the napkin calculations to figure out how i’d get there.

20 radishes a week:

I’m not going to go over the full plan here, but instead go through some higher level examples to demonstrate the thought process and the insights and implications that arose from changing my approach.

So to reach 20 radishes a week, with an assumption of potentially 6 months of grow period in the year left, I would need approximately (20 x (6 x 4.5)) = 540 radishes.  I’d need to plant 20 radishes a week, possibly 25 a week to compensate for those that don’t germinate, I probably don’t want to be planting every week so perhaps i’ll plant 50 radishes every 2 weeks.

Hang, on, where am I going to plant them.  If each radish needs to be planted 3 to 5 cms from each other, in rows 10 to 20 cm apart, then 50 radishes is going to need an 1/8th of a square meter.    Every 2 weeks i’m going to have to plant 50 radishes on 1/8th a square meter, and I should be harvesting that amount too every week.

Seeing as they take 4 to 6 weeks to mature, and can probably stay in the ground for up to 8 weeks, then i’ll need at least 4/8th of a square meter constantly sowed. I’m going to double that because I figure it gives me more flexibility as I can always harvest more if I need, and adjust my sowing rate up to compensate.

This is doable, but only now that i’ve figured out the details of how to micro scale production for this one crop.  It completely changes the experiments that I need to conduct.  And firstly, I need to get a much more aggressive planting schedule.

The Plan:

When I expand this type of thought experiment to Salads and Tomatoes, it becomes clear that my challenge here is not going to be the subject matter expertise of how to grow a good salad, or how to grow a healthy juicy tomato.  The actual challenge is going to be how do I do produce at scale!!??? (in this case scale doesn’t actually mean scale …it’s relative to what I have done before).  And  learn the subject matter expertise along the way to get healthy juicy results.

This now becomes a logistics challenge.  And I do love a good logistics challenge.

This is the game plan at this stage:

  • Radishes, testing several types in outdoor in ground bed.
  • Salads testing 2 types, aquaponics and in ground bed,
  • Tomoatoes testing several types, in aquaponics (flood & drain grow bed) and in raised bed


Full circle learning:

The great news is that this method will also allow me to learn fast because i’ll have a constant flow of germinating plants to mess up and learn with.  Inevitably, this constant sowing strategy is going to provide me with the focus and velocity to go deep and master some focused areas.  Looking forward to it.

Wish I could do the same at work.








6 x sprouts in 4 days

Overall, the system seems to have jumped up a gear.  The fish seem well and are feeling strongly.  The plants have taken a torn towards looking better. 

Another good sign, the 6 lettuces I sowed 4 days ago have all sprouted. (although you have to look close to see them all).

With the system more balanced, and with the addition of a fan, I’m hoping these sprouts start off with better stems. 

Last note to self:  to have a more substantial source of  vegetables, I need to figure out a sowing schedule so that I’m constantly growing more plants to replace one’s I hope to one day harvest. 

Time to figure out planting

When  I first planted my plants for the aquaculture it was pretty cold, and it was 4 weeks until i got fish.  

As a result many of my young plants have issues with their stems, and generally doesn’t stay of life strong.   I’m therefore getting rid of the runts (and feeding them to the fish) and resowing a few lettuces. 

The new seeds above are:

3 x laitue verte maraîchère 

3 x laitue rouge Grenobloise 

Planted in floating raft 2, 30th March, water temp 24.1 C.