-
-
Notifications
You must be signed in to change notification settings - Fork 34
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[VOTE] Accept SLEP000 #64
Conversation
I'm not against merging early, but I'm strongly against a process that will prevent additional feedback and comments from being adequately expressed and addressed #30 (comment). The proposed workflow in this SLEP has limitations that have been voiced by multiple core devs (#30 (comment), #30 (comment), #30 (comment)), but they are not addressed nor acknowledged in this SLEP proposal. For these reasons I'm -1, sorry. |
+1 for accepting the SLEP the way it is (merge Draft PR early). I think we can improve the review/discussion without setting it in stone SLEP 000. One idea could be to maintain an open META discussion issue for each SLEP in Draft state to centralize the discussion with a maintained list of open discussion topics in its description with potential links to sub-issues and sub-PRs related to the draft SLEP for instance. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I acknowledge that I am not 100% happy with the workflow proposed but I don't have a better proposal or counter-proposal at the time of the vote.
I give therefore my +1 hoping to provide some amendment in the future to improve the situation.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
While I acknowledge that we could probably improve upon the details of this proposal, I believe that:
- It is a net improvement over what we currently have
- Good design is iterative and (unlike with some other SLEPs) we don't have a burden of backward compatibility here.
Hence I am +1 and approve. Thanks to everyone involved!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I find it a bit weird to accept a Process SLEP:
- that introduces explicitly "A Process SLEP describes a process surrounding scikit-learn [...]; they require community consensus." and that does not reach a consensus.
- that introduces explicitly a workflow for SLEPs (merge => follow-up discussions and reviews => vote), and that goes directly from merge to vote.
But I agree it is important to move forward (which is what this very SLEP aims to improve). Maybe we can mark this SLEP as "Provisional" ?
Counting the votes, it's 10 in favor (votes above plus myself), and one opposed. The vote satisfies 2/3 majority of cast votes. Therefore this motion (accepting SLEP000) is passed per our governance. |
This PR is the place for us to vote for SLEP000: https://scikit-learn-enhancement-proposals.readthedocs.io/en/latest/slep000/proposal.html
It was discussed in depth in #30, but no unanimous consensus was achieved there.
Therefore here's the vote for us to see if it passes our majority. We require 2/3 of the cast vote (no quorum) to pass the SLEP.
The vote closes on 4th February 2021.