It’s usual for us to spend a great deal of time on a specific paper, especially when we try to understand the current state of research in a field. Thus the efficiency when reading is critical for us to keep current. Related blogs and papers about how to improve reading efficiency I’ve read are listed below.
To start with, all of them told me that you should determine the order of the reading. The first thing to do is to gain a general idea about the paper, which means carefully reading the title, abstract, introduction, and conclusions as they contain the summary and structure of the study from which we can directly get the key points. By doing these, we should be able to answer the below questions:
This information is adequate for us to decide whether this study is worth further attention. Sometimes we will find the paper is not in our research area or have poor results. In that case, we should save time for more related ones. However, if this paper interests me a lot, we can go on next step.
Then we care for the assumptions and design parts. This process can be hard especially when we have never heard of some background knowledge and assumptions they have used in their design. In my opinion, the only thing we can do is to spend time learning about what we know less but may play a role in the paper we are reading, sometimes the references of the paper can help, or simply search for it on the web. For example, I’ve recently read a paper about a protocol design to thwart selfish-mining attacks on the blockchain, then I may have to find some material about self-mining attacks if I’ve never heard of it. During the design part of this study, we can focus more on the core algorithms and take notes to highlight the key points of the design so that you can soon grasp the content of the paper when you read it next time. Details such as mathematical proofs can be ignored if unnecessary because it may take much more time to understand them than reading a total paper. At the end of this process, we should figure out how they solve it and why it solves.
What I’ve never considered before I read these reading suggestions is that we should not assume that the authors are always right. After at least one time of the total pages, we should ask some appropriate questions. Are the assumptions the authors make reasonable? Do they have limitations? Are there simple solutions the authors do not seem to have considered? At the same time, we should also think positively. Do these ideas have other applications or extensions that the authors might not have thought of? Can they be generalized further? Are there possible improvements that might make important practical differences? If possible, compare their work to other related works to understand what they actually make for the whole field.