Episode 74: Building on Top of CHAOSS Software
CHAOSScast
Building OS's Compass: Data Analysis Platform
In this chapter, they discuss the goal and process of building OS's Compass, a SUS platform based on matrix model definitions from chaos. They explain how users can analyze data without setting up their own tools infrastructure and collect data, and highlight the integration of GITI data source.
Thank you to the folks at Sustain for providing the hosting account for CHAOSSCast!
CHAOSScast – Episode 74
On this episode, our host Georg Link kicks off the discussion, introducing a stellar lineup of panelists including Sean Goggins, Yehui Wang, Mike Nolan, and Cali Dolfi. The topics discussed today are the CHAOSS software, Augur, and GrimoireLab, and the different applications built on top of this software. The panel members discuss the projects they are involved in, such as the Augur project, OSS Compass, and Project Aspen’s 8Knot. Then, we’ll delve into Mystic’s prototype software, aiming to transform how academic contributions are recognized and valued. The discussion dives deep into the role of CHAOSS software in open source and community health, talks about Augur and GrimoireLab projects, ecosystem-level analysis, and data visualization. Press download now to hear more!
[00:00:58] The panelists each introduce themselves.
[00:03:03] Georg explains the origins of CHAOSS software, particularly Augur and Grimoire Lab, and their development. He dives into Grimoire Lab’s focus on data quality, flexibility, and its identity management tool, Sorting Hat.
[00:05:55] Sean details Augur’s inception, its focus on a relational database, and its capabilities in data collection and validation. Georg and Sean recall Augur’s early days, focusing on GitHub archive data, and its evolution into a comprehensive system.
[00:09:28] Yehui discusses OSS Compass, its goals, the integration of metrics models, and the choice of using Grimoire Lab as a backend. He elaborates on OSS Compass’s ease of use and the adoption of new data sources like Gitee.
[00:14:16] Mike inquires about the handling of the vast number of repositories on Gitee, and Yehui explains using a message bus and RabbitMQ for both data handling and parallel processing. Sean clarifies that Gitee is a Git platform similar to GitHub and GitLab, and OSS Compass is the metrics and modeling tool.
[00:15:29] Cali asks about the visualization tool used, and Yehui mentions moving away from Kibana to front-end technologies and libraries like ECharts for creating visualizations, which is an Apache open source project.
[00:16:29] Cali describes 8Knot under Project Aspen built in Plotly Dash and Repel, focusing on mapping open source ecosystems using Augur data. She emphasizes the data science approach to analyzing open source communities and the templated nature of 8Knot for easy visualization creation by data scientists.
[00:20:19] Sean comments on the ease of adding new visualizations with Dash Plotly technology in 8Knot. Cali adds that new visualizations can be easily made an that 8Knot is connected to a maintained Augur database but can also be forked for specific community and company needs.
[00:2342] Georg underlines the importance of ecosystem-level analysis, especially for software supply chain security. Cali shares the goals of analyzing ecosystems to understand relationships between projects, influenced by Red Hat’s interests in investing in interconnected communities.
[00:26:30] The conversation shifts to Mystic, and Mike describes it as a prototype software integrating both GrimoireLab and Augur, with the goal of better integrating these projects through development.
[00:27:30] Mike outlines Mystic’s goal to serve as a front-end to date collection systems, with a specific focus on the academic community’s contributions to technology research. He envisions Mystic as a tool for academics to measure community health and impact of their projects, aiding in tenure and promotion cases.
[00:30:52] Yehui asks about integration of Grimoire Lab and Augur within Mystic and the selection of components for the solution. Mike explains the early stages of integration and the plan to combine data collection services from GrimoireLab into Augur to support undergraduate student development.
[00:32:30] Mike details research on Mystic, including interviews with faculty from various departments to understand their digital collaboration and artifact creation. He aims to develop generalized models of collaboration applicable to multiple data sources, allowing systems like Mystic to support diverse academic disciplines.
Value Adds (Picks) of the week:
- [00:36:26] Georg’s pick is focusing on the slogan, “One day at a time.”
- [00:37:12] Cali’s pick is doing a Friendsgiving this week.
- [00:38:08] Sean’s pick is the launch of the tv show ‘Moonlighting’ from the 80’s.
- [00:38:49] Yehui’s pick is riding his bike to work which is peaceful for him.
- [00:39:52] Mike’s pick is attending The Turing Way Book Dash.
Panelists:
Georg Link
Sean Goggins
Michael Nolan
Cali Dolfi
Yehui Wang
Links:
Building an open source community health analytics platform (Mystic)
Special Guests: Cali Dolfi, Mike Nolan, and Yehui Wang.
The AI-powered Podcast Player
