One more year, we attended to our traditional visit to FOSDEM and celebrated our classic GrimoireCon, which this year has become into CHAOSSCon! We want to share with you our experience with a recap from CHAOSSCon and also from the talks some of us gave at FOSDEM.
This year’s CHAOSSCon had a very tight schedule, as there were 16 talks and 6 workshops! Manrique, our master of ceremonies and CEO, started CHAOSSCon 2018 with one of the biggest novelties: as you may know GrimoireLab is now part (and one of the founding members with Bitergia) of the CHAOSS project, by the Linux Foundation.
The first talks were related to the current state of CHAOSS and GrimoireLab. One of the highlights was the talk about Diversity & Inclusion metrics, while the second bunch of talks focused on use cases from both industry/large-scale projects (OW2, Mozilla, The Linux Foundation, CROSSMINER) and academia (University of Mons) from the usage of metrics in these open source communities to the challenge of dealing with identities information.
Later, it was time for the lightning talks where we presented some of the new tools coming to our GrimoireLab family:
- Hatstall: a tool for managing identities information within your project, which provides a Web UI where you can operate merging and affiliation information.
- Bestiary: a tool to configure in a simple way your analytics scope using a Web UI.
In these lightning talks we could learn about the work from Open Source Metrics team at Twitter; about Prospector, a project from Red Hat to provide a set of metrics of projects for success, sustainability and vibrancy; and finally about GHData, a Python library that provides data related to GitHub repositories using GHTorrent database.
For the first time in GrimoireCon/CHAOSSCon history, we had two parallel tracks for workshops:
- Track 1 – Analysts and Developers: Jesús gave and intro to set-up GrimoireLab platform from scratch and how to analyze the data we obtain from it.
- Track 2 – Using GrimoireLab Dashboards: Alberto and Daniel explained the features of the dashboards and how to create new visualizations and other dashboards. Later, David and Daniel showed some use cases solved with GrimoireLab capabilities, like social network analysis with David’s network plug-in for Kibana.
And after CHAOSSCon…
As usual, the cover our visit to FOSDEM in our three usual strategic positions: our humble booth full of owly presents, our table-headquarters at the cafeteria and also distributed among rooms giving some talks around!
This year, there were three talks given by Bitergians at FOSDEM:
David Moreno’s talk about his Network plug-in for Kibana
With a room packed of people, David gave his talk about the Network plug-in he developed for Kibana at the Community Dev Room. David explained which kind of networks can be produced with his plug-in and how to configure them. Also, he showed some cool use-cases obtained with the GrimoireLab platform, like the dashboard with data from Marvel API, containing comics-related information (authors, illustrators, publication dates, etc.) and building cool networks to display this information.
Jesús M. González-Barahona’s talk about GrimoireLab platform
Jesús started his talk about GrimoireLab, the free, open source software
platform to produce software development analytics with a quick overview of GrimoireLab’s architecture and some of the features from each of its components. In the end, Jesús made a demo of the whole platform producing a dashboard: everyone was able to try the full platform with a single command to run a Docker image with the full platform up and running!
Valerio Cosentino’s talk about Perceval, our tool for data-retrieval
Valerio stated in his talk the aim of the GrimoireLab tool Perceval, giving as an example some of the questions we may want to solve while we are analyzing a project, such:
- Are there new contributors?
- How many bugs were fixed past month?
- Does our gender diversity has increased lately?
To answer these questions (and more), we need to gather data from many different data sources, and this is where Perceval comes to play. Valerio detailed how Perceval obtains data and how it structures it, making a lot easier to produce new backends for Perceval to support more data sources (Currently, it has support for 25+ different ones), in addition to the possibility of using Perceval as a independent program or a as a Python library.
…with all the owls (and GrimoireLab T-shirts too) flown away…FOSDEM 2018 came to an end!
See you next year!
Leave a Reply