The best type of Problems are Root Node

Root Node problems are the most important problems

“We cannot solve our problems with the same level of thinking that created them” - Albert Einstein

I think a lot about solving problems and I’ve come to discover that not all problems are equal. Some problems are more important than others. There are some problems which when solved lead to exponential progress and essentially take a field to the next level. This essay attempts to talk about those essential problems. Problems I call Root Node problems.

What are Root Node Problems?

By definition, root node problems are the type of problems which have multiplicative effects across the entire system, they compound and lead to exponential benefits. They are bottlenecks that restrict the progress of a broader domain and when solved compound downstream.

I think the best way to understand the concept of root node problems is by providing examples of existing root node problems and then giving examples of non-root node problems.

• The problem of hallucinations in Large Language Models is a root node problem. When it’s solved it opens up new avenues of use cases for LLMs like AI-based psychologists, AI writers, etc

• Another problem is that of plasma instability in Nuclear Fusion. If that can be solved then we’d be a lot closer to Nuclear Fusion as a sustainable power source.

• Another interesting problem is creating a centralized data hub where health data can be accessed by researchers to train models aimed at solving problems.

• Every 2 minutes a woman dies in childbirth. Ensuring safe childbirth is a crucial problem to be solved.

Now let’s give examples of non-root node problems.

• Building a dating app. The world doesn’t need another swipe left and find love app.

• Building a crypto-currency. The world doesn’t need a crypto-currency right now.

• Building a Twitter clone. The world doesn’t need another twitter/X.

It is really important to note that Root Node Problems are not relegated to fields of study. They can be scaled down to individuals (the root problems of an individual) or scaled up to the root node problem of a society. But we’re going to stick with fields of study because they are the most impactful part of this spectrum and quantifiable(the root node problems of society and individuals tend to become philosophical).

What makes up Root Node problems?

We’ve answered the question of what root node problems are, now let’s attempt to understand what makes up root node problems. Here are 3 defining characteristics of root node problems.

Foundational

Root node problems are foundational by nature and they serve as the basis for future progress. For example, if hallucinations in Large Language Models aren’t solved then the use cases for chatbots(like ChatGPT) will essentially be relegated to non-sensitive tasks.

Blocking

Root node problems are major obstructions to progress in broad fields. They are essential barriers which delimit progress in domains. In other words, progress can’t be achieved if they aren’t solved. For example, if the plasma instability problem isn’t solved then nuclear fusion is going to forever remain unfeasible.

Multiplicative

Root node problems by nature compound and have exponential downstream effects. For example, here’s a trick multi-step question - how many Mozarts and Einsteins do you think we can get if we can stop women from dying every 2 minutes during childbirth? and how much better would the world be?

How to Discover Root Node Problems

The best heuristic to make use of when searching for root node problems is to ask the experts, by “experts” I mean those who are at the cutting edge of their field. These people usually have very good idea (at the very least the best idea) of what the most important problems are in their fields and the best way your resources can be expended.

Asking experts is a really good heuristic because root node problems by nature are the crucial element stalling future progress in certain fields and by nature can be difficult to spot for someone without knowledge esoteric to the field in question.

It is also really important for you to carry out independent research to find out the best problem for you to solve, based on variables like the skill level available to you, financial resources, the current state of research on the problem, and the current state of existing technology.

It is equally important to map out dependencies required for stated problems to be solved. Dependencies in this case are (but not relegated to) government permissions. Bottlenecks like these usually occur in the heavily regulated healthcare and educational sectors. A good example of this is the use of CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) to replace the Hemoglobin cells in adults (hemA) with those in fetuses(hemF) to replace the sickle-shaped cells with healthy ones, it took quite a few years for the solution to be licensed for clinical.

End Note

At the premise of this essay, I stated that not all problems are equal and some problems are more important than others. I’m hopeful I’ve explained faultlessly the logic behind that. On that note, I believe it’s really important that we as a society work to invest the bulk of our resources (financial, intellectual, technical) into solving them.

To achieve that from a truly foundational level we must indoctrinate the importance of these problems from the university level.