Open thesuperzapper opened 3 years ago
Hey @kubeflow/project-steering-group, here is my proposal which was spawned out of the discussion in kubeflow-discuss: https://groups.google.com/g/kubeflow-discuss/c/WlAtv8yBwGE
The main purpose of this issue is to track progress on the roadmap.
CC all current Working Group leads: @kubeflow/wg-automl-leads @kubeflow/wg-deployment-leads @kubeflow/wg-manifests-leads @kubeflow/wg-notebooks-leads @kubeflow/wg-pipeline-leads @kubeflow/wg-serving-leads @kubeflow/wg-training-leads
I believe the primary google contact for this work is @theadactyl
Hi there, as mentioned in the thread on kubeflow-discuss, this isn't something the PSG is working on at the moment. For now we'll be prioritizing work on the items in our roadmap, including items which would be required for any TSC to meaningfully administer central processes: https://github.com/kubeflow/community/blob/master/psg/ROADMAP.md This is in line with PSG scope to "establish the procedures and process in order to smoothly transition to a diverse structure, as defined by the group".
A point about the TAC -- the TAC was not disbanded, but instead saw attrition in 2020 as former members were no longer able to participate to the same degree in the project (due to job changes and/or shifts in their companies' focus). That is why jlewi promoted the creation of decentralized working groups that would have minimal "project-wide technical decisions" requiring central resolution, and @Bobgy has been working to facilitate decentralized Working Group operations.
I'll leave this issue open (backlog) and will provide updates on roadmap and timeline when possible. As I mentioned before, please focus on opening issues against concrete technical or procedural pain points you or other Working Group members are experiencing.
Thanks for pulling this together @thesuperzapper!
One thing that I think would be incredibly helpful, is for us to identify a set of things between now and establishment of the TSC (and break them out into their own individual issues). A small list for this, I can think of is:
Feel free to augment - but let's get at what a TSC will provide in the interim as a parallel effort.
Since we have shipped Kubeflow 1.3, I think forming the elected "Technical Steering Committee" should be priority number 1 for the current Google-run "Project Steering Group".
The current ROADMAP isn't working towards this goal, but I think Google can start to make process by:
@theadactyl what do you need from the community to support this? (For example, if resource/time is an issue, we can write up the TSG election process for the PSG to approve)
The need for the Technical Steering Committee came up in today's community meeting with regards to the creation of the release team for 1.4 and how it's members will be chosen. I think it is important to perform the steps Mathew outlined in the previous comment for the upcoming release.
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
I strongly believe that this issue should be pursued further, in the interest of having a neutral and elected governance structure for the Kubeflow project.
/lifecycle frozen
Goals
Background:
Back on 2019-05-02, the community accepted a governance structure proposal: https://docs.google.com/document/d/1HKB662Ju6URVdzw0Neq9OFnxM6RMXFu4Dd04q9E3kGk
This governance structure defined a future group called the "Technical Steering Committee" which was intended to be a diverse and community-elected leadership group for Kubeflow.
In order to facilitate the creation of the "Technical Steering Committee", the governance structure defined the following interim governance groups. (NOTE: these groups were not intended to be permanent)
Working Groups:
Technical Advisory Council:
Project Steering Group:
Steps:
Clearly we have gone a bit off-track during 2020, but I believe we can quickly move to form the "Technical Steering Committee" by undertaking the following actions: