Here are one-per-collection polls for all the currently included collections.
The polls have two options:
Choose “Yes” if you think the collection satisfies the new inclusion criteria and should stay in the Ansible community package.
Choose “No” if you think it should be excluded.
Let’s keep it open for 2 weeks, and I’ll close it on 2026-11-18.
It’s a community public poll. The results will be informational.
I’ll adjust the PR correspondingly.
Then we could conduct the second poll in the steering committee if needed.
Please vote.
UPDATE: There are categories-related comments below the polls. As I would avoid editing the posts with open polls, please read the comments at the bottom of the thread.
I voted no on storage related collections because they all seem vendor specific. But I also feel like some storage related stuff would be a good addition so I’m a bit conflicted.
Hopefully someone with more familiarity can express an opinion on any of them.
I think it can be seen as both. Also community.sops should likely be in the same categories. (SOPS is similar to Ansible Vault in that you can use it to encrypt secrets in your repository.)
IIRC if we include community.general won’t that automatically mean we need community.library_inventory_filtering_v1 as it’s a collection dependency for community.general.
I went through the list and voted for most of the collections (for some I’m unsure). Generally I voted for keeping a collection if it is not too niche and manages something which is open source and doesn’t require paying some vendor (by money or data) to use it. While I’m not sure whether it’s a good idea to select collections only based on that, I think it’s a general guideline that should play an important role in the decision.
I think that, if the purpose of the ACP is to give users something to start with, that’s a very small set of tools. If they are moving from that newbie level to something more serious, the very first thing they need to handle is ansible-galaxy install. Therefore, I voted no for almost everything. My old and beaten up 2 cents.
Joined to make this point but glad you have, for the vast majority of these things, I voted based on ‘if I were getting to know ansible from scratch again, coming from my floss tools first background, what do I need to successfully run a homelab or set up some management over some cloud boxes and rental VPSes, day 1, before I figured out what ansible-galaxy is or how to make a docker runner / ephemeral execution environment.’.
We know that ansible-core is for ‘the devs’ and ansible needs to be ‘enough to start with’, but I tried to leave in packages that will just be assumed to be necessary for broad uses cases (back in the day it was everybody has to install paramiko for windows etc. etc.)
I agree with @russoz. Me specifically, I work mostly with the Check Point collections. When I first started, the immediate thing I noticed was that the ACP package was already outdated and I still had to do ansible-galaxy install. My first thought back then was “well that was a waste of effort for the ACP folks…”.
Later, I did work with the Azure, then VMware, collections and again had the same issue and thoughts (plus, “huh? There are TWO VMware collections!?”). So yeah, vendors should do their own work if they’re gonna be serious about it.
I don’t mean to be specifically “vendor-shaming”, but… well, maybe just a wee bit.
Thank you, Steering Committee, for making this poll available!
Apropos of nothing, comunity.grafana and community.hashi_vault are on my list of possibly unmaintained collections in ACP 15.0.0. Looks like there haven’t been new releases since last year, at least as far as I can see…
I’ve just closed the polls - thank you all very much for spending some time on voting and discussing the stuff! I’ll move the not-overwhelmingly-no collections to the “keep list” in the PR