"Too much information, just like too little information, is an obstacle to understanding. In other words, modern media is hindering our understanding with an overwhelming flood of information."
Before We Begin#
I've wanted to write an article about how I manage "information" for a long time. Recently, I saw Randy's share on Twitter and read the book "Building a Second Brain." I also reconsidered how I should manage information.
The purpose of good information management.
- How to quickly find solutions to problems encountered in work or life
- How to generate more interesting ideas
Input#
Currently, my main methods of information input are the following:
Information Stream#
- GitHub
- YouTube
- Jike
I follow many great developers, designers, and some interesting product official accounts on Twitter. From these follow lists, I can see their latest viewpoints and what new technologies and products have emerged. I use Likes as a bookmark; whenever I see content I'm interested in, I like it. Additionally, I've created some Lists, which serve as a categorization feature; the major categories I've established so far are AI, FE, and Web3.
GitHub also has an information stream — "For you," which many people might not use much. From here, I can see which repos the developers I follow have starred, and it also recommends some recently popular projects. I categorize the starred repos, which you can see here.
On YouTube, I mainly browse some subscribed channels. The content is quite broad, covering technology, life, and entertainment. I often discover high-quality content here, such as the Web3 Mooc from UC Berkeley, street photographer Kai W, and table tennis blogger Adam, among others.
Jike is a great Chinese community, with decent content on Web3 and AI, and some active independent developers and investors. However, the latest tech news is still slightly behind Twitter.
The main feature of the information stream is — algorithm recommendations, but it is a double-edged sword; sometimes it brings surprises and helps discover more interesting people. On the other hand, it can easily lead to "information overload," with users refreshing the app over and over again.
RSS#
RSS subscriptions are another common way I obtain information. They are mainly used to subscribe to some tech blogs, but there are also some other themed sources. For design-related content, Behance, Design-Milk. For tech-related content, 9to5Mac, The Verge. For Web3-related content, Weekly in Ethereum, Foresight news, etc.
Compared to information streams, I believe RSS has the advantage of higher article quality — because the sources are filtered by myself before being added to the reader. I regularly refresh the reader to obtain this high-quality information input.
Reading Books and Listening to Podcasts#
If I want to learn a subject more systematically, I believe reading books is still the best way to learn. Of course, finishing a professional book is quite challenging. In most cases, Google (or perhaps now directly asking ChatGPT) or reading documentation can solve the problem. Very few people would spend a lot of time understanding the ins and outs of a problem.
The advantage of podcasts is that they provide first-hand information. For example, if I want to understand Sam's views on the future development of AI, listening to related podcasts is more credible than reading articles summarized by related media.
CODE Method#
The CODE method is the core content of the book "Building a Second Brain." Below are some of my own practices.
- C (Capture) -> Capture resonant information
- O (Organize) -> Organize and categorize
- D (Distill) -> Highlight the essence
- E (Express) -> Showcase the results
For the "Capture" part, it mainly includes the several methods mentioned above, and I also use the Web Clipper plugin to clip web articles into my Yuque knowledge base. Then, I organize in the Yuque knowledge base — through folder categorization. Furthermore, I write summaries in Obsidian to complete the "Distill" phase, and finally, I write blogs, tweet, and post on Jike to showcase the results.
Completing these steps completes the entire process of the CODE method. Of course, in practice, adjustments should be made according to one's own situation, as everyone uses different tools. For me, I feel that I haven't done enough in D and E, which is the "output" part.
Thoughts#
Good information management means that when learning new technologies or solving previously unencountered problems, there's no need to start from scratch; instead, it builds on an existing knowledge base. In the future, large language models combined with personal knowledge bases may make the process of building a second brain simpler. However, regardless of the method, what we should focus on is "where the information we've seen will go in the future." I want to conclude with a quote about TypeScript that I came across before.
As you write TypeScript, you need to think about where you want your contracts to be, and what needs to be done in order to meet them.