Recognizing AsyncAPI OSS contributors

[AceTheCreator]

Description

We are currently working on a new addition to the website called community, which will reside at https://www.asyncapi.com/community. While we've made 70% progress, we still need a critical section which is known as the contributor's section.

This section will be used to recognize the people who inspire and contribute to our OSS community. I'm basically using this thread to track the progress and manage discussions related to this feature.

Required features

We have some ideas outlined to make this more interesting, including the following.

• Contributors Badge (Thinking of having a maximum of 5 badges that can be achieved)
• Amount of Pull requests
• Amount of Issues
• Amount of project contributed to
• Contributors' public promotions (Special shoutouts)

Current Progress

Implementation

To make this a reality, we need a way to retrieve contributors' data, and yes, we chose to opt-in for gharchive instead of GitHub API because of the following reason stated by @derberg:

We have over 60 repos (63), the rate limit is 60 requests, and they reset every hour.

• you need one call to get a list of repos
• you need 63 calls to get the list of contributors
• you need a call per contributor (over 1000 requests) to get their details so you can:
  • check their forks
  • check issues they created (not sure it is doable, probably another call per repo and then do user matching)
  • check PRs they created (not sure it is doable, probably another call per repo and then do user matching)
    And I did not even include the counting of requests increased due to pagination.

Optimistically you need 1100 API requests to build a list. If we have a 60 requests limit, it is a GitHub Action running for about 18h.

For the above reason, we've decided to go with gharchive, which is a project that records the public GitHub timeline, archives it, and makes it easily accessible for further analysis. But now, this imposes a new challenge which is we get all archived GitHub data for a specific date range. Then we have to do further data cleaning and a couple of analyses on the data to meet our needs, probably by using Big Query or some other technology. Luckily for us, we found someone in the community whose strong field is Data Analysis. @thulieblack is currently playing with the data to see what is the best way to overcome this challenge

Design

We have some design being done by @Mayaleeeee, which can be found here. Feel free to share your thought or probably help enhance it :)

We also need help with some design ideas for these Badges, please if you're a designer, feel free to come up with something sweet 😄

Please feel free to share your suggestions.

I'll keep this document updated as we make progress. Stay tuned ✌🏾

CC @derberg @thulieblack @alequetzalli @Mayaleeeee @asyncapi/tsc_members

[AceTheCreator]

We had a few exciting conversations about the implementation of this feature in the community meeting that happened on the 21st of March, 2023. Here's the recording https://www.youtube.com/watch?v=tKYyL_DiLds&t=600s. Feel free to share your ideas in the comment below :)

[thulieblack]

Update:

It's possible to get an individual person's number of commits, timestamp, and more details but it comes at a heavy-duty price. Running a query from 2016-present requires over 17 TB talk less of exporting and storage.
However, I did run a sample for the month of February 2023 which took about 340GB
Here are the results although I couldn't extract them in full.

Costs

According to Bigquery Pricing, for running queries you get 1TB free every month and afterward pay an amount of $5 per TB. For storage, you get 10GB for free every month then pay $0.01 per GB.

Since LF has done something similar for their Insights dashboard although they focus on overall data, I went ahead and posted on their forum page to ask if it would be possible if they could share the dataset they use for building these dashboards or point us in the direction of how we could access it.
Will give an update once they respond.