Adapting and Thriving: My Quickstatements 3.0 Internship Journey


Hello, welcome back to my blog. This will be a short progress report on the first half of my internship and how I've had to strategize to adjust the timeline of tasks.
To recap, I'm working on a project called Quickstatements3.0, a tool used to batch edit Wikidata items. You can read all about it here.
At the start of my internship, my objectives were clearly defined: to master the tool's command syntax and create comprehensive, user-friendly documentation for both new and existing users.
I created a timeline to achieve this, centered around completing three steps in the following order:
Understand the Tool: To use the tool, one must understand the command syntax. Quickstatements has two command syntaxes: the v1 command syntax and the CSV command syntax. Understanding both intimately is fundamental to writing easy-to-digest documentation.
Test the Application: This involves using my command syntax knowledge to develop complex datasets for thorough application testing..
Write the Documentation.
However, two weeks in, I realized a need to revise my approach. It became apparent that tackling steps 1 and 2 concurrently—understanding the syntax, creating test datasets, and testing the application—would enhance my comprehension of the command syntax and structure my workflow better.
Doing this took longer than I expected, by the end of the first month, I had only completed the dataset creation and testing for the v1 command syntax.. It made sense to begin writing the documentation alongside testing, rather than waiting to finish testing both syntaxes.
I spoke to my mentors about adjusting the timeline to reflect this change, I received approval to modify my timeline, resulting in the following revised steps:
Step 1, v1 command syntax: Understanding the syntax, create the test dataset, test the application and write the Documentation.
Step 2, CSV command syntax: Understanding the syntax, create the test dataset, test the application and write the Documentation.
This task grouping allows me to focus on and complete each section before moving to the next, adding structure to my workflow.
I have recently finished the documentation for the v1 command syntax and will begin the CSV command syntax in week 7, marking the second half of my internship. With this established workflow, I anticipate it will take two weeks to complete the dataset creation and documentation for the CSV command syntax. The final weeks will be dedicated to review submissions and implementing any necessary corrections.
I am excited for the second half of my internship and cannot wait finish up creating the documentation.
Subscribe to my newsletter
Read articles from Onitcha Nyerhovwo directly inside your inbox. Subscribe to the newsletter, and don't miss out.
Written by
