Communitas Vinculum - Part 2
This post is regarding the last community bonding meeting. The official coding period starts on the 27th of this month. Only a few days are left, and preparations are in full swing.
Today, on the 24th of May, my mentors and I had a meeting on #grimoirelab
.
This meeting was very important since it discussed several key points for the start of the coding period.
Agenda of the meeting
- Decide a fixed time for weekly IRC meetings with mentors
- Decide on a different approach to reference implementations
- Plan for the next few days
Summary of the meeting
-
A Fixed Time for Weekly Meetings
Meetings will be held on fridays at 13:15 (CEST) for me. Nischith, a fellow GSoC student for the Grimoirelab project, will have a meeting with at 12:30 CEST. The meetings are to be held consecutively so that Nischith and I can attend each other’s meetings. The duration will be approximately 45 mins. -
An approach for Reference Implementations
Jesus, my mentor, proposed a great idea regarding the reference implementations. It goes like this: create a class for the analysis of reference implementations with several methods, the most important being a method which would calculate a single value for the metric. Some otherclass MetricCompute(Period, Source_Code_criteria): # generate a value for the metric method1: compute(self, data) # generate a value for each period: like one per week/month method2: compute_time_series(self, data, period) # generate charts / graphs regarding the metric optional: generate_graphs(self, data, ..)
methods can be added depending on the metric to create charts or graphs. This notebook will be used to generate a script, which would be tested against some sample Perceval JSON data. Then a user can import this as a module to run it on their own data. An issue will be opened soon in the wg-evolution repository for re-implementing the Code_Changes-Git metric.
-
Plan for the Next few days
We decided that it would be best if I created a sample reference implementation in my project-tracker repository so that it would be possible for others to give feedback on the effectiveness of the new strategy.
The log for this meeting can be found here.
The next meeting will be on May 31st. It will be the first meeting of the official coding period. I will write a post following that next week.