An alternative extra credit opportunity is to read the paper: Code Code Evolution: Understanding How People Change Data Science Notebooks Over Time

Code Evolution: Understanding How People Change Data Science Notebooks Over Time Previously notebooks (IPython, SciPy) were used as code editor, scientific calculations and for distributed computing. But now this trend is changing. Now a days the notebooks are becoming one of the essential tools that helps the data scientists to perform their works more accurately. A survey of Data Science Notebooks: The survey was performed over 2,574 Jupyter notebooks which were selected from 60,000 Jupyter Notebooks and sensemaking spectrum regression model was used to get the result of sensemaking activity within individual notebooks. Four types of groups of sensemaking shifts are considered in the meta-analysis which are exploration to exploration, exploration to explanation, explanation to explanation, explanation to exploration. Outcome of The Survey: The result also shows that most of the shift happens in explanation-to-explanation category which indicates that data scientists may prioritize clarity and reproducibility during sensemaking within notebooks, consistent with prior work. The notebooks are being updated by the number of annotations, correct data sets and markdown cells as per the requirement of the data scientists. But still, we need to improve the notebook features in several areas in future. Improvement Areas: Traditional notebooks are good for software development, they are still missing some of the core features. For example, we need mechanisms that can be able to track actual non-linear and iterative practices of data scientists and there should be some way so that notebook can calculate the position of a notebook document automatically. Apart from that we need better UI/UX for collaboration and workflow-oriented approach for the notebook and developers and data scientists are continuously working on it. Future of Data science Notebooks: By analyzing the trends of the notebooks in GitHub we can predict that the future notebook will have high agility, better integration, and collaboration with the remaining data beyond doubt. Example: Jupyter Notebook is an open-source web application which becomes the most widely used notebook interface for data science now a days. The non-linear overflow of this notebook makes it more interactive so that it helps the users to jump between coding and the notebook and becomes an essential part of conducting data science.