Reading paper can be fun but recalling papers are not. If you do not synthesize something out from reaidng papers, it is likely that you get nothing out of it after a few months you read it unless you really love the papers. However, I would say there are a limited number of papers that you can love, which is much lower than the papers that you have to read if you would like to become a researcher. In that sense, logging papers that you read is essential. You will recall more and more papers as you become senior but up to that point, you will simply read it and get nothing other than the good feeling.
Therefore, I will ask everyone who does research with me to maintain an annotated bibliography (or something equivalent) per a project. If you have any preference (or existing methods), the annotated biblography will be a simple Google Docs document that contains a collections of paper and the length of which grows over time. In the long run, I would recommend a bibliography software such as EndNote, (which is free for Virginia Tech students as of now) as long as you can share it with me. A stack of papers with manual annotation is pretty good but it is not a good way for collaboration. However, other types of collaborative systems works e.g. Github, a Wordpress, Medium, etc. I would recommend something that requires minimal transaction costs.
Each entry on one paper should include following information
- Title of the paper
- Authors of the paper
- the most official url of the paper (e.g. ACM DL)
- Brief summary of the paper (2~4 sentences)
- Method (what they did and how did they validate it?)
- Self reflection on the paper
- how it relates to your project?
- in which purpose would you cite it.
In addition, it is great if you annotate the paper (pdf/paper) so that you can quickly review how you liked the paper. In that sense, it makes sense to also store the pdf somewhere. EndNote or Google Drive, both of which are free for Virginia Tech. students. For annotation, personally, I tend to simply highlight the sentences that resonates in me. Also, it is good to highlight keywords and sentences that can summarize each paragraph.
When reading a paper, these are the general questions to that I would stick to.
- What is the problem that they are attacking?
- What is the research contribution of the paper?
- What is their research method?
- What can be generalized?
- What is a missing knowledge gap that they try to fill in?
- How does it relates to my research?
- What are the findings that helps me do whatever I can do?
- How my research is different from the work and similar to my work?
For more information about annotated bibliography, please check out the following links: