FI Learning

For learning with practice. Posts are not private and could end up on Bing.

Understanding Levels of Detail: Zooming Trees In and Out

Project steps are organized in trees (that’s how they work whether you know it or write them that way or not). All trees can be zoomed in to a higher level of detail by looking at just a sub-tree. And they can be zoomed out to reduce detail by hiding the descendants of some nodes. This is a crucial skill to understand for trees in general in the abstract, and also to be able to apply to real examples with big or small trees. It needs to be practiced until it’s intuitive, not just something you can do when you stop and think about it. You need to see this stuff automatically and instantly so your full attention is free for harder stuff.

This is easier to show with a video. There are many ways to represent trees. You can use bullet point lists with nesting (indented sub-lists). You can write trees on one line by using parentheses around each list or sub-list and quotes around each item. But the easiest way to understand trees is with diagrams. You can draw trees with boxes for each item (node), lines showing the connections, and have it organized spatially in rows.

Practice zooming projects in and out. Make a basic 2-level project tree (root is project name and children are steps). Then add additional levels above and below. Do this multiple separate ways. And redo the original tree for the same project: write different steps for the same activity, and then do more levels above and below. Compare everything and see what seems realistic and useful to you. Imagine doing it and what would be useful or bad (too much detail or too little detail). Imagine explaining it to others and what would be useful. Look for errors. Actually do some projects and see what following the steps is like. Try turning your brain off, not using common sense or common knowledge, and only following the steps, and see how well or badly they work. You shouldn’t expect it to work perfectly that way but you can see how broken it is – some lists of project steps break more than others which shows they were making more assumptions and leaving more out.

Some common types of errors to look for:

  • Parent/child mismatch. The child is not a detail or sub-part of the parent.
  • Different levels of detail as siblings.
  • Skipped steps. Project won’t work due to missing parts. Example: Step 1 drink a coke. Step 2 get paid by the recycling place. There were no steps about finding a recycling place, saving the can, taking the can there, etc. When stuff just kinda appears out of nowhere, or teleports around, then steps are being skipped.
  • Unnecessary or irrelevant steps.

All projects are trees, whether you understand that or not. There are always different levels of detail that could be written down. You always make decisions about what levels of detail to plan at. You must practice looking at other levels of detail in order to develop good judgment about what works well. Just like you must practice thinking about and writing out projects in a variety of ways (e.g. 10 different sets of steps to accomplish the same thing) in order to develop good judgment about what steps to include and how to make them fit together well.

Project planning

Getting better with project steps is one part of project planning. Here’s a quick outline 

  • goals (w/ success criteria)
  • resources
  • steps
  • risks
  • resources used by steps and resources potentially used by risks

A resource is anything useful. People generally think about resources too narrowly and ignore most of them. Some types of resources:

  • time
  • mental energy
  • physical energy
  • money
  • help from friends
  • help from workers
  • raw materials
  • tools
  • skills
  • knowledge
  • machines
  • electricity, gas, batteries
  • muscles
  • land

Physical objects are resources, but more specific categories like “tools” or “money” are generally more useful to think about. And note that my list is categories (types) of resources. What you actually list for your project is e.g. “hammer” or “$50”.

When budgeting resources for projects, it’s worth considering how much using a resource consumes it, or whether you can easily reuse it. Time and money are consumed when used, but hammers and carpentry training can be used repeatedly. Help from friends can be used more than once but not too often. Being able to use a resource for more projects makes it more valuable to acquire.

There are many, many types of resources. You must always make judgment calls to decide which to consider and budget for your projects. To get good at this, you must practice. That means including many different resources in your practice projects and seeing which ones end up being important. Practicing also means leaving resources out of project plans and seeing if anything goes wrong.

What to do

Either practice or, if you have trouble figuring out what to do and how to do it, ask questions and work on figuring it out yourself too.

The best way to ask questions is to try it yourself, and start your own learning process, and then share info about what you tried and why you’r getting stuck. Questions like “I don’t get it – please explain how to do it.” are not useful. My post already tried to explain stuff. If you want personalized help you have to get started at least a little bit and then provide personalized info about what you did and what the trouble is. You have to try to solve your own problems and ask for help with the ones you fail to solve. You can ask for help quickly, like after a few minutes, but it should be framed in terms of you solving your own problems, and taking action yourself, and where that learning process is struggling. Ask for a boost with what you’re doing yourself, not for someone else to think or act for you.

If you just go days without practicing or posting, because you stopped, that’s on you. You’re choosing not to be involved. The occasional break is fine. But if you don’t practice stuff then you’re never going to make progress. Reading without practice isn’t good enough to learn much.

Comments & Events

Anne B
In the video, around 6:00, it says that a good project has around 2 to 5 steps. Why 2 to 5? Is it because that's an easy number to keep in mind at once?
Elliot, Fallible Ideas
0-5 children is a general rule of thumb for making trees. It makes them a reasonable medium, rather than too tall and skinny or too short and wide.
Anne B
So it makes a tree easier to read if it's closer to a square shape. Does it also make thinking about the layers of steps easier if each layer is 2-5 steps? I think yes.
Elliot, Fallible Ideas
it's a content issue, not presentation. if something has 50 parts you're not dividing it up well and things aren't organized. those parts should be organized into groups.
Anne B
I'm thinking about what you wrote here:

Goldratt said that even for complex multi-billion dollar projects like building an offshore oil drilling platform, the project plan shouldn't have more than 300 parts max. (IIRC he was talking about PERT charts.)

A project plan with 300 parts sounds big and scary. But if it's organized in a tree shape, with each node having only 2-5 children, then it's not so scary. If you're making a plan like that, you probably already have knowledge about how to do lots of the sub-parts.
Justin Mallone
Good video Elliot! 🙂

During the video I thought about things that seemed related to me, like integration and the crow epistemology stuff from Objectivism. I enjoyed thinking about those connections.

On reflection, I realized that while I have been using trees more lately for some purposes and finding them useful, I don't often fold them up! That's kind of interesting! I didn't even know the keyboard shortcut for doing so in MindNode which really tells you how little I do I'm going to try folding up the trees I make more, think about them at different levels of folded-up-ness (and really, think about the conceptual categories that correspond to different levels of foldedness and whether they make sense or not) and see how useful I find that. It seems like a way to get more out of trees. This is not a very ambitious goal but I'm making and using trees already so it seems like an easy one to accomplish.
Justin Mallone
I recently discovered that Atom lets you fold sections of code. That seems related to folding up parts of a tree. Both some code and part of a tree represent some knowledge within a bigger structure. Anyways this is pretty cool
Justin Mallone
AnneB said: 

A project plan with 300 parts sounds big and scary. But if it's organized in a tree shape, with each node having only 2-5 children, then it's not so scary. If you're making a plan like that, you probably already have knowledge about how to do lots of the sub-parts.

Anything with 300 disorganized parts sounds kinda scary. Even 300 disorganized letters is a big mess without them being grouped into words and sentences. 
Elliot, Fallible Ideas
300 parts is really low for a multi-billion dollar project with well over 300 ppl working on it. the point was to reduce the number of parts from e.g. 3000.
Elliot, Fallible Ideas
oil rigs, by default, should sound way scarier than 300 parts