Why we can’t recall what we read
This is Part 1 of the “Improve knowledge conversion with book tracking” series. In this series, I will explain why we should build a tracking system to help converting what we read into actionable knowledge. In Part 2 and Part 3 I will explain the technical details of building one.
Have you spent a lot of free time reading books, only finding out later that you can’t recall anything? This problem is very prevalent and has less to do with whether you understand the subject well. Rather, it’s more related to how our memory retains information and the learning style we adopt.
When we first encounter new data (or ideas), they are stored in our short-term memory. If we don’t process and output them within a specific time frame, we run the risk of losing them (imagine them as data cached in our machine’s RAM, instead of SSD for long-term storage). By post-processing (e.g., filtering, analyzing, building references to what we’ve learnt before) and practicing what we’ve learnt (writing, producing a video, teaching others, applying it to a project, etc.), we convert them into concrete knowledge. Moreover, our brain transfers such knowledge to long-term memory.
Thus the text we read from books are merely data points if we just stop there. What a pity if we consider the time we spent browsing book catalogs, checking out the ones we like, and actually allocating a slot from our busy daily schedule to read them, only to let all the text vaporize a few weeks later! If we borrow the conversion funnel view from marketing, we’ll notice that the vast amount of campaign budget (i.e., our time in acquiring data) we expended has translated into zero purchase (useful knowledge), an absolutely horrendous investment.
Improve knowledge conversion with better techniques
Graphics credit: NTL
In fact, the learning pyramid proposed by the NTL in the 1960s suggests something similar, that passive learning such as reading/viewing videos/attending lectures results poorly in memory recall, while active learning such as practising projects and especially teaching others yields a much higher knowledge absorption rate. While the actual percentages are now up to debate and the current take is to suggest a mix of activities for better balance, teaching others is widely considered as a good tool to enhance learning. When you have to explain newly learnt concepts to others, it forces you to reflect deeply on what you’ve learnt, clarifying your thought processes and helping you to see the weak points in your knowledge base. Your listeners might ask you questions that go into the realm you’ve not explored, forcing you to be humble with what you really don’t know. What might seem obvious before now becomes questionable, prompting you to do additional research. No wonder Richard Feynman also proposed people to explain in plain language what you’ve learnt to kids or grandmas, via the so-called Feynman Technique. Jo Boaler, an education professor in Standard University and author of the book Limitless Mind, also suggests using a multidimensional approach and learning with others to improve our learning potential.
As I jot down my notes and arrange them in a structural way after completing a book, I’m reminded again of the once-foreign concepts I came across while reading. This reinforces my learning and challenges me to be honest with what I really don’t understand. If I have the aim of explaining the book to a friend from the start, I become even more focused while reading, always looking for key points that should be communicated to someone new to the book. Making a few slides for mock presentation is another good exercise post-reading.
During the COVID19 pandemic I began to checkout more books from the library. With the proliferation of systems such as Overdrive offered by your local library and Kindle Unlimited, it becomes even easier to binge read for pleasure, and lose track of the actual time I devote to reading. If, borrowed from the popular term, that each of us is our own life’s CEO, then I start to question if my time is really well spent, judging by the return I get from each activity I engage in. Do I become wiser, more balanced, and productive after reading this book? Do I get to understand another valuable perspective from the author’s experiences? I want to be more deliberate in measuring whether my input is yielding good returns. I’m not now to notetaking, as I’ve been using various journaling methods (ranging from paper to Evernote, OneNote and various) to capture book notes for a long time. But I’m never quite satisfied, since they are decentralized and hard to piece together. Most importantly they don’t track my time investment. So last year I began to tackle this issue in a more systematic way.
Why I built my own knowledge tracking system?
My objectives are to
- own my data, so I can do whatever I want with it (analytics, doing CRUD (create, read, update, delete) operations
- implement a solution quickly with low to no cost
- allow me to track my time and books
- remind me to test my memory retention and practice what I’ve learnt, to accelerate the conversion of data into knowledge, and through continual practice, into wisdom
- stop relying on 3rd party vendor apps to access my data
I have been an avid user of the Chinese book review site Douban Books since 2008. Every year they would generate a year-end report to recap the books you’ve read. While I thoroughly enjoy the quality of community reviews there, it owns your data like many other social sites. There is no way to export my data to chart and measure my progress.
Goodreads is better as it allows you to export your data as CSV files. However, they closed down their API access in 2020, which means I will have no way to tap into live data for dashboarding or doing CRUD operations programmatically.
All these commercial offerings are useful in recording the books read and storing your notes, but they all force you use their own apps to interact with your data. They also lack a way to track time investment and memory recall. All these prompt me to create my solution from scratch.
What tools did I employ?
Capturing tabular data with records of multiple fields is easily done in spreadsheets. So I started by using Google Sheets to track all the books I’ve read. One thing I want to accomplish is to minimize the time for book entry (author name, title, published year, language, genre, description, cover image, etc.). To accomplish this, I use an Apps Script in the sheet to look up and pre-fill book information using Google Books API. The sheet also contains other columns that store my notes, status, next steps and calculates how much time I spend on each book. It also calculates how much time has elapsed since I’ve completed a book, and will send a reminder to prompt me into recalling whether I could still remember, or better yet, have been utilizing the key takeaways.
Next, we need to create a way for us to enter and edit data. If you’ve edited a spreadsheet on your phone before, you know how unpleasant the experience is. The cells are just too small to single out even when you’re using a native app. So the mobile Sheets app is out. If you only want to do quick adds from your mobile device but will edit your notes in Sheets, then a Google Form (associated with the sheet) is a fast solution to implement. Unfortunately it doesn’t allow us to read, update and delete existing records via its interface so we need to employ an alternative.
So I implemented a mobile-friendly UI using Google AppSheet. AppSheet is a modern “no to low code” development environment where you can quickly connect multiple data sources and implement functionalities to create web and native mobile apps with a simple interface. In other words, my Google Sheets worksheet acts like a pseudo database whereas the AppSheet app acts as a web frontend.
In Part 2, I’ll share with you all the steps I’ve taken in building such a system.