The theme of this article is how to conduct research work and how colleagues can assist each other. It mainly consists of some thoughts from my previous work on technical research, which may not be suitable for everyone. (I haven't read any related books either)
(Now here is a picture)
Clarify the Research Direction#
Before conducting research, it is essential to clarify the goals and roadmap of the research, and communicate these plans and expectations with relevant colleagues. During this process, organizing thoughts is key. Discussing with colleagues is a good way to clarify thoughts.
The implementer of the research is usually not the same person as the one who proposed it, so confirming the needs of the proposer is important. Before starting the research, you can try to briefly summarize the research content based on your understanding and confirm it with the proposer. Unclear requirements may lead the research in the wrong direction, as in the case of a GitHub Action research where, due to unclear requirements, the focus was placed in the wrong direction.
Focus on Key Research Points#
The key to research is to "determine which information obtained is valuable to us" and focus on narrowing down the scope. For example, when researching a book, it is not necessary to list all the points but to extract the parts that are valuable to us. Otherwise, a mere chronological account of the research is meaningless.
In startup companies, we often do things we are not good at or in new areas, making it difficult for new colleagues to conduct research because they may not be able to judge which information is valuable to our work. That's why it is not recommended for beginners to directly read technical books, not that they shouldn't read, but they should not read aimlessly.
What the Research is About#
Understand what problem the technology solves. The technologies we research are generally aimed at solving practical problems, so it is important to clarify the boundaries of problem-solving.
Compare various solutions, there are always trade-offs, understand why a particular solution was chosen over another.
Identify the usage scenarios of the technology, as scenarios can help determine if the technology is suitable for us.
Construct Classic Cases#
Research involves many abstract concepts, such as workflows, steps, and jobs in GitHub Action, which may be difficult to explain in words. In such cases, constructing classic demos and integrating concepts into them can be effective. Demos need to be concise and include more features.
Others (Regarding the Relationship Between Popular Science and Research)#
Popular science and research have a similar point, which is "explaining unfamiliar knowledge to others in understandable terms". Here, I recall the author Wang Jie from the column "Science Has Stories", where he mentioned that "the work of popular science writers is to extract real, accurate, and objective information from the ocean of information, process it in their own language, and then popularize it to make it easier for everyone to understand and absorb". Or like why articles by Ruanyifeng are so popular, because his articles are actually quite similar in nature to popular science articles.
In conclusion, technical research requires clarifying the direction, focusing on key points, explaining the problems solved by the technology, constructing classic cases, and effectively conveying complex concepts in understandable language. I hope these personal experiences can help you in your research work.