Last year I launched the Anti-Book Club, my own take on the tradition of book clubs.

The idea is simple: instead of everyone in the group reading the same book, duplicating time and effort, we each read a different book on the same topic. Then each person summarizes the book they read, and I compile and share the summaries with everyone who contributed.

The goal is to divide the effort of reading a large collection of works, and thus to conquer a complex topic in a small fraction of the time it would take one person to read all of them. Along the way, we produce a valuable collection of succinct book summaries that we can easily refer to in the future.

This experiment relied on a few key tools:

A standardized template that everyone used to summarize their book
A technique I’ve developed for saving the main points in a book without losing their surrounding context, called Progressive Summarization
Tools for exporting the highlights from ebooks (such as Bookcision, Readwise, Clipping.io, and built-in export options for Kindle and iBooks)
The digital note-taking program Evernote

None of these tools are essential for reading and summarizing a book, of course. But they helped us make our summaries more consistent, and thus understandable, for anyone who might read them in the future.

I summarized the full guidelines for running your own Anti-Book Club in these official instructions, in the hopes of spreading this practice to any group that needs to learn a large body of knowledge in a short time.

Using these guidelines, we completed two month-long rounds of the Anti-Book Club last year. 59 participants summarized 59 books related to note-taking, visual thinking, personal knowledge management, and extended cognition, all drawn from my reading list for the book I’m writing on Building a Second Brain.

I originally shared this archive only with those who contributed in order to provide an incentive, but now I’ll share the Evernote notebook that contains all the summaries with all Praxis members:

Access the Archive

I’ve spent the last year slowly working my way through these summaries, and now I’d like to share with you what I learned from the experience.

5 Things I Learned From the Anti-Book Club

1. Reading well-designed book summaries is an intoxicating experience

I knew from past experience that progressively summarizing a note is an inherently rewarding experience.

Because each layer of summarization takes only a fraction of the time compared to the one before, it feels like you are rapidly building momentum with each layer. For smart people reading about interesting topics, more insights coming at a faster pace is highly rewarding.

Reading a progressively summarized note is also rewarding. Because all the work of finding the key points and digesting the core message has been done in advance, you can capture the gist of a note in a very short period of time. I found reading the summaries took me only about 10-20 minutes each.

What I didn’t expect was the effect of reading a series of well-designed summaries in one sitting: I can only describe it as intoxicating, like drinking a highly distilled liquid made of pure insight. Instead of long stretches of boring explanation punctuated by brief moments of insight, like most books, I was consuming only the densest nuggets of paradigm-shifting ideas.

This effect was probably stronger for me because all the summarized books were drawn from my reading list, which means I pre-selected them for interest. But even for newcomers reading about the topic for the first time, I believe this collection of summaries has a lot of value.

Here’s why: with the surrounding text around each main point left in context, each reader can calibrate how much attention they pay to any given paragraph or idea. If a given highlighted passage seems obvious, there’s no need to dive into all the details around it. But if a passage seems counter-intuitive and novel, the full explanation is just a glance away in the surrounding text.

I found myself doing this from summary to summary, and even within the same summary. As I read, I was constantly modulating the amount of attention I was paying and the level of detail I paid attention to, using my intuition as a guide. I believe this is more in line with how human attention works – flitting from object to object seeking novelty – rather than how we’re taught to read in school, absorbing each word in uniform and predictable order.

I found that the way I “read” these summaries was very different than how I read books. My eyes remained slightly unfocused, searching for highlighted passages and key words, moving up and down and sideways, going back and re-reading parts, jumping ahead in anticipation. It felt much more like examining a painting in a museum, prioritizing resonance over thoroughness.

2. Only a small percentage of books are relevant at any given time

This learning confirmed what I suspected from the beginning: that only a small percentage of these books would be directly applicable to my current project.

Out of 59 books, specifically chosen for their relevance to the book I’m writing, I found that only 9 were directly applicable. These included:

This is quite an astounding outcome if you think about it. It calls into question at least my ability to correctly judge which books will be most useful. Which is a risky proposition considering each one represents a 5-10 hour commitment.

I don’t think this says anything about the quality of the books I did or didn’t use. I think it’s more a question of timing. Many books would have been helpful, say, a year before, when I was in “divergence” mode and exploring any kind of interesting idea. But now I am in “convergence mode,” driving toward a published book as fast as I can.

I think of it as a “frontier of knowledge” for any topic one is exploring. As you read and learn, you cover more ground and the frontier is pushed out further and further. A source that a year before might have been highly insightful, might now be completely obvious. One that in the past was too challenging, might just have been waiting until you are ready. If only books that lie right on your current frontier are truly useful, that implies that their number will be quite small.

But this is exactly why summarized notes are so valuable. If you read a book that you’re not quite ready to put to use, you can package it up in a note that will remain preserved for whenever that day comes. And if you read something that is too basic, you can package it up as review material for the future. Or even put it to use in other ways, such as using it to bring others up to speed more quickly.

3. Book summaries fell into 4 categories based on how valuable they were

I quickly found that, as I finished reading each summary, I would make one of four decisions about what to do with them:

FOR DEEPLY INSIGHTFUL, VALUABLE BOOKS: READ THEM IN FULL MYSELF

These were books that seemed to contain a profound idea or insight that were crucial to my work, but couldn’t be reduced to bullet points. I would add the book to my reading list.

FOR BOOKS CONTAINING NUMEROUS VALUABLE POINTS: DUPLICATE THE NOTE AND PUT IT IN A PROJECT NOTEBOOK

Most of the 9 books above fell into this category. They followed a thread that was relevant to my project at numerous points, and I knew were worth revisiting. I would duplicate the note, and move it directly to the most relevant project notebook for later review.

FOR BOOKS WITH A FEW USEFUL QUOTES: COPY THOSE QUOTES INTO A NEW NOTE

Other books had a few choice passages which captured the essence of the idea well. I didn’t need the full-length notes to refer back to, so I simply copied these passages into a new note, and moved them to the most relevant project notebook. I knew if I needed to find where they came from, a quick search would lead me back to the full notes.

FOR BOOKS WITH INTERESTING POINTS, BUT NOT RELEVANT TO ANY CURRENT PROJECT: TWEET THE QUOTE OR IDEA

Many other books contained interesting or insightful points, which were often highlighted by the summarizer. But if they had no relevance to any current projects or interests, the only thing left to do was to share them with my audience.

Each of these actions represents layer 5 of summarization: translating, distilling, triaging, or sharing the summarized notes to make them more discoverable for my future self and for others. In the process, I also absorbed them more deeply myself, and created value for my followers.

4. Progressive Summarization is a partially subjective process

The decisions of what to highlight, and at which layer, proved to be fairly subjective. I noticed some passages distilled to layer 3 and even 4 that I wouldn’t have thought twice about. And others I considered crucial only made it to layer 1.

This is, of course, to be expected. If compressing a text was a completely objective process, it would be done by computers. The value in Progressive Summarization is that it makes this subjectivity a little less risky. If someone makes a different decision about what to save than you would have made, it isn’t lost. It’s just preserved in a different layer than you would have chosen.

That said, there are some best practices emerging about how to make summarization decisions that are most likely to be useful to future readers. I’ll identify those best practices for future rounds and synthesize them into a short video.

5. This process produces a different kind of relationship to “books I’ve consumed”

This learning emerged over a period of months, as the ideas I gleaned from these books rattled around in my head. I clearly hadn’t read them in the usual manner, and definitely didn’t have an intimate familiarity with them.

It was more like I’d “consumed” or “downloaded” these books. Like Neo does with kung-fu in The Matrix – the knowledge is there, and it took only a fraction of the time to gain, but requires some working out in the dojo before it’s fully integrated.

I think this distributed reading process changes our basic relationship to the books whose main ideas we’ve consumed in summarized form – they become more like tools or commodities, prized more for their functional utility than for the subjective experience they impart.

That may sound a little mercenary, but I think it’s a good thing: you’re still free to go back and immerse yourself in the full experience if you want. But for those books not worth reading in full, you’ve saved yourself a lot of time. Especially for non-fiction books whose purpose is to convey a message, not tell a story, I think distributed reading has the potential to expose us to a wider variety of ideas while saving time.

This changed relationship had a few other interesting effects. I noticed a distinct lack of attachment to the ideas I encountered. I realized that dedicating 5-10 hours to a book could sometimes lead to the sunk cost fallacy: overvaluing information just because I’d spent a lot of time to acquire it. I was much more willing to pass over ideas that were good, but not great, knowing that something better probably lay just around the corner.

I’ve been surprised how satisfied I’ve been with my understanding of these books. I’m usually very perfectionistic and self-critical in my reading, always fearing that I’ve let something important fall through the cracks. Now I realize that maybe that fear comes not from my dissatisfaction with my note-taking skills, but from the huge upfront investment required for so little value.

It’s just possible that we read books not to read books, but for the morsels of insight they give us. Lacking a way to gain these insights directly, we’ve had to settle for consuming huge tomes that often disappoint. This new form of collaborative, distributed reading could change the equation, deconstructing books into their most useful ideas, like a car being stripped for parts.

Why does that matter? Because every time we reduce friction in acquiring information, a revolution happens. When the barrier to entry is lowered, more people use that information for more purposes. In the Internet Age it might seem like all the friction has already been eliminated. But an abundance of information has led to a poverty of attention.

It’s about time for us to pool and share that attention, organizing it into something greater than the sum of its parts. The new frontier of collective knowledge is not huge databases or more advanced technology, but simple and proven techniques for synchronizing human intellectual labor.