Metrics

At OpenQ, we aim to create opinionated metrics to facilitate a better understanding of the information we retrieve from GitHub repositories or users. To enhance comprehension of each metric, we provide a list of algorithms utilized. We welcome any feedback you may have.

Repo activity score

This algorithm calculates a repository's activity score based on several factors:

1. Commit Count and Time Frame: It considers the number of commits made to the repository and the duration over which these commits occurred.

2. Consistency Scores: It takes into account the consistency of these commits over time, represented by a set of consistency scores.

3. Calculation Steps:

   • It first calculates the average number of commits per day by dividing the total commit count by the number of days.
   • It calculates a combined consistency score by averaging the provided consistency scores.
   • It prepares a score using a mathematical formula involving logarithms and weighting factors based on commit frequency and consistency.
   • The prepared score is capped at a maximum of 100% to prevent it from exceeding this limit.
   • It ensures a minimum activity score of 10% if at least one commit was made in the past 6 weeks.
   • Negative scores are prevented, with the minimum score being set at 0%.

Overall, this algorithm aims to provide a comprehensive measure of a repository's activity level, taking into account both the quantity and consistency of commits over time.

User activity score

This function calculates the activity score for a user based on the following parameters:

• commitCount: The total number of commits made by the user.
• days: The duration, in days, over which the commits occurred.
• combinedConsistencyScores: An array of consistency scores representing the consistency of the user's commits over time.
• log: A logger object used for error logging.

The algorithm follows these steps:

1. Calculate Commit Frequency: It computes the average number of commits per day by dividing the total commit count by the number of days.

2. Compute Combined Consistency Score: It calculates the combined consistency score by taking the mean of the provided consistency scores.

3. Prepare Score: Using a mathematical formula involving logarithms and weighting factors, it computes a prepared score that accounts for commit frequency and consistency.

4. Score Adjustment:

   • If the prepared score exceeds 1, it limits the activity score to 100%.
   • If the prepared score falls below 0.1, it ensures a minimum activity score of 10% if at least one commit was made in the past 6 weeks.
   • Negative scores are prevented, with the minimum score set at 0%.

Overall, this function aims to provide a comprehensive measure of a user's activity level, considering both the quantity and consistency of commits over time.

Popularity score

This function calculates the popularity score of a repository based on the number of stars it has received. Here's how it works:

1. Input Parameters:

   • repository: Information about the repository, including the number of stargazers (users who have starred the repository).
   • log: A logger object used for error logging.

2. Handling Missing Repository:

   • If the repository information is missing, the function returns a default score of 0.

3. Calculating Simplified Score:

   • The function assigns a simplified score based on the number of stars:
     • 0 stars: Score of 0
     • 1 to 10 stars: Score of 1
     • 11 to 20 stars: Score of 2
     • And so on, up to 10 stars: Score of 10

Overall, this function provides a basic measure of a repository's popularity based on the number of stars it has received, with higher scores indicating greater popularity.

User reputation score:

This function calculates the reputation score of a user based on various factors, including their contributions to reputable projects, popularity score, and interactions with issues. Here's how it works:

1. Input Parameters:

   • userData: Information about the user, including their top repositories and issue comments.
   • commits: Information about the user's commits.
   • userPopularityScore: The popularity score of the user.
   • log: A logger object used for error logging.

2. Calculating User Reputation Score:

   • It calculates the user's contributions to reputable projects using the getUserReputableProjectsContributions function.
   • It computes the user's interactions with issues, represented by the total count of issue comments, normalized to a scale of 200.
   • The reputation score is then calculated using a weighted sum of these factors:
     • 50% based on contributions to reputable projects.
     • 40% based on the user's popularity score.
     • 10% based on interactions with issues.

3. Logarithmic Scaling:

   • The calculated score is scaled logarithmically using the logWithBase function with a base of 3.5 to ensure a balanced scale.

Overall, this function provides a measure of a user's reputation based on their contributions to reputable projects, popularity, and interactions with issues, with higher scores indicating a higher level of reputation.

Repo reputation score:

This function calculates the reputation score of a repository based on the contributions made by its assignable users. Here's how it works:

1. Input Parameters:

   • repository: Information about the repository, including assignable users and their contributions.
   • log: A logger object used for error logging.

2. Handling Missing Repository:

   • If the repository information is missing, the function returns a default score of 0.

3. Calculating Reputation Score:

   • For each assignable user:
     • It calculates the reputation score based on their contributions to repositories.
     • Contributions are determined by factors such as the number of stars and forks received by the repository and the number of commits made by the user.
     • The reputation score is calculated as the sum of the reputation scores of all users divided by the total number of assignable users.

4. Final Score Adjustment:

   • The preliminary reputation score is adjusted using a logarithmic function to ensure a balanced scale.

Overall, this function provides a measure of a repository's reputation based on the contributions made by its assignable users, with higher scores indicating a higher level of contribution and reputation.