18:01:24 <samccann> #startmeeting Community Working Group
18:01:24 <zodbot> Meeting started Wed Apr 12 18:01:24 2023 UTC.
18:01:24 <zodbot> This meeting is logged and archived in a public location.
18:01:24 <zodbot> The chair is samccann. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
18:01:24 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
18:01:24 <zodbot> The meeting name has been set to 'community_working_group'
18:01:27 <cyberpear> o/
18:01:33 <anwesha[m]> Hello
18:01:41 <samccann> @room Meeting time!
18:01:46 <samccann> #chair anwesha
18:01:46 <zodbot> Current chairs: anwesha samccann
18:01:49 <oranod> o/
18:01:51 <gotmax23> .hi
18:01:52 <zodbot> gotmax23: gotmax23 'Maxwell G' <maxwell@gtmx.me>
18:02:01 <samccann> #chair cyberpear Don Naro gotmax23
18:02:01 <zodbot> Current chairs: Don Naro anwesha cyberpear gotmax23 samccann
18:02:12 <samccann> #info Agenda: https://github.com/ansible/community/issues/679 / Topics: https://github.com/ansible-community/community-topics
18:02:26 <samccann> #topic Updates
18:02:35 <samccann> #info Ansible 8.0.0.a1 in the works today, sorry for the delay. Includes ansible-core 2.15.b2
18:02:48 <maxamillion> .hello2
18:02:49 <zodbot> maxamillion: maxamillion 'Adam Miller' <maxamillion@gmail.com>
18:02:49 <samccann> might actually be out already?
18:02:57 <gotmax23> anwesha: I acked it. Thanks!
18:02:58 <samccann> #chair maxamillion
18:02:59 <zodbot> Current chairs: Don Naro anwesha cyberpear gotmax23 maxamillion samccann
18:03:04 <Leo[m]> Hi all
18:03:15 <samccann> #chair Leo
18:03:15 <zodbot> Current chairs: Don Leo Naro anwesha cyberpear gotmax23 maxamillion samccann
18:03:28 <samccann> #info new docsite is live at docs.ansible.com. Very few people choosing to go to the old site so seems a success!
18:03:39 <samccann> #topic community.grafana and grafana.grafana merger CLA requirement
18:03:43 <cybette_> o/
18:03:51 <samccann> #chair cybette
18:03:51 <zodbot> Current chairs: Don Leo Naro anwesha cyberpear cybette gotmax23 maxamillion samccann
18:03:53 <samccann> #info discussion: https://github.com/ansible-community/community-topics/issues/212
18:04:00 <samccann> #info last comment suggests merger won't happen.
18:04:08 <gotmax23> I'm a bit unsure about the new docsite. I still can't figure out how to click through to access the community section/collections index
18:04:20 <samccann> gotmax23:
18:04:45 <samccann> can you open an issue for that https://github.com/ansible/docsite
18:04:54 <samccann> I admit I've had a similar problem
18:05:27 <oranod> gotmax23: you know, that could be on me. there might even already be an issue for that here: https://github.com/ansible/jinja-docsite/issues/78
18:05:27 <gotmax23> meh, I didn't phrase that well. That made it sound like I didn't like the new site at all. I just meant that one part could be better displayed 😃.
18:05:44 <samccann> coolness
18:06:20 <oranod> there's defo still some work to do. I'll get cracking on that one though, if it's the same as what you were referring to.
18:07:20 <samccann> #info no active votes at this time
18:07:45 <samccann> #topic Ansible roadmaps
18:07:57 <samccann> #info - created a project roadmap so we track alpha/beta etc releases - https://github.com/orgs/ansible-community/projects/7/views/1
18:08:01 <Leo[m]> oranod: +1, had the same issue. But was on mobile, had the double burger top bar and dismissed it for later. Then forgot. 😅
18:08:13 <samccann> That's to help us stay on track with the alpha/beta/rc etc
18:08:13 <gotmax23> wait, should we finish the grafana thing?
18:08:22 <samccann> oh yeah got ahead of myself
18:08:43 <gotmax23> nah, it's my fault for interrupting :)
18:08:54 <samccann> So the last comment on the graphana issue was that they couldn't comply so thus won't merge the two collections
18:09:00 <samccann> https://github.com/ansible-community/community-topics/issues/212
18:09:06 <gotmax23> right
18:09:08 <gotmax23> that makes me sad
18:09:41 <acozine> o/
18:09:48 * gotmax23 waves
18:09:51 <samccann> #chair acozine
18:09:51 <zodbot> Current chairs: Don Leo Naro acozine anwesha cyberpear cybette gotmax23 maxamillion samccann
18:10:12 <samccann> is there anything else we can/should do about the graphana issue?
18:10:12 <acozine> sorry I'm late - notifications REALLY don't work when you don't have your client running ;)
18:10:16 <samccann> s/graphana/grafana/
18:10:20 <samccann> hehehe
18:11:25 <Leo[m]> samccann: If they are not merging, and community is not being maintained, that means it's going to be deprecated... Right?
18:12:17 <gotmax23> are you referring to https://github.com/ansible-community/community-topics/issues/212#issuecomment-1478353759?
18:12:35 <samccann> gotmax23: yes
18:12:51 <samccann> are there next steps (like starting deprecation in ansible 8 or 9)?
18:13:11 <samccann> or asking grafana to maintain both for now?
18:14:38 <samccann> oh reading the discussion a bit better...
18:15:00 <remindbot[m]> @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting
18:15:43 <cyberpear> fwiw, I think openstack generally requires CLA but dropped the requirement for the ansible collection (but only after I complained)
18:15:56 <acozine> I can't find anywhere where they say explicitly why they want to require a CLA
18:16:29 <gotmax23> I'm not sure if that means that community.grafana is completely unmaintained and should be removed
18:17:24 <acozine> yeah, it seems like the person maintaining the community collection is the same person or one of the same people maintaining the corporate collection?
18:17:29 <gotmax23> I think they offer an "enterprise license" / paid version
18:18:38 <gotmax23> acozine: I don't think it's the same person. rrey is maintaining the community one and
18:18:39 <gotmax23> ishanjainn the grafana one.
18:19:02 <acozine> ah, but rrey can't do it any more, right?
18:19:18 <gotmax23> > I have not been able to give time to the collection maintenance for several months now and having Grafana involved sounded a good opportunity to avoid its death.
18:19:57 <gotmax23> if community.grafana is going to die anyways, then I guess grafana.grafana might as well absorb the content
18:20:01 <samccann> so thinking out loud - if community.grafana passes the 'test' as unmaintained, seems we do need to start the deprecation process. And maybe grafana swoops in before it's removed to solve it all, or maybe it goes away like other unmaintained collections
18:20:25 <acozine> it seems like either way we end up with one collection with a CLA - either the work gets merged together into the collection with the CLA, or the community collection gets no maintenance and only the collection with the CLA continues to be usable
18:20:46 <samccann> gotmax23: wouldn't that be erm.. a license violation? If grafana.grafana takes the code, don't they have to take the license?
18:21:09 <gotmax23> I'm upset about that, but I'd rather have the code the community wrote maintained by someone instead of completely dying
18:21:19 <gotmax23> samccann: not if they get the community members who wrote it to sign the CLA
18:21:20 <acozine> agreed
18:21:47 <acozine> I hope we can find maintainers for it in future
18:21:48 <samccann> gotmax23: yeah do not like the feel of that for sure (community people having to sign a CLA so their work doesn't die)
18:22:40 <samccann> but there's nothing we can do to stop something like that (a corp asking a contributor sign away rights).
18:23:11 <samccann> So, are we agreed we need to start the deprecation process on community.grafana?
18:23:23 <samccann> or do we need more peeps chiming in? I know a few are out today
18:23:32 * gotmax23 is mostly speculating about the reason for the CLA. I'm not sure why they need it for the collection. Sometimes, it's just a corporate policy for all of the the company's OSS
18:23:45 <samccann> I'm also happy to just add this as a comment in the issue and see who responds
18:23:53 <gotmax23> yeah, that sounds like a good idea :)
18:24:30 <samccann> done
18:24:39 <gotmax23> obviously, we should see what rrey thinks about that
18:25:09 <samccann> #topic not allowing any more CLAs
18:25:27 <samccann> #info discussion: https://github.com/ansible-community/community-topics/issues/221
18:25:27 <gotmax23> I'm still onboard with this :)
18:25:30 <samccann> seems like a good next topic :-)
18:25:46 <gotmax23> yup!
18:26:07 <gotmax23> does anyone have major objections?
18:26:12 <gotmax23> I'm thinking I should start a vote
18:26:59 <samccann> for the cluless... what is DCO?
18:27:26 <gotmax23> https://en.wikipedia.org/wiki/Developer_Certificate_of_Origin
18:27:59 <samccann> cool thanks
18:28:00 <cyberpear> demonstrated as `Signed-off-by` on a git commit, usually
18:28:07 <samccann> and yea, I'm +1 for starting a vote on this one.
18:28:12 <gotmax23> cyberpear: right
18:28:15 <Leo[m]> I'm not sure a hard no is a good idea without checking why they did it for example, and if there are other valid cases and how to handle those
18:28:31 <samccann> is there an 'easy' way to see what other collections might have a CLA already?
18:28:52 <Leo[m]> they cited the ASF CLA as source for ex. not sure it includes the patent clause there, but ASF has an individual CLA and corporate CLA for ex
18:29:35 <gotmax23> basically, it's a simple agreement that confirms the contributor has the right to submit the code they're contributing
18:30:00 <remindbot[m]> @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting
18:30:07 <samccann> Leo: I guess it depends on our philosophy for the Ansible package.
18:30:51 <samccann> like if 99% of the package doesn't have a CLA, would a contributor notice that THIS particular one she's about to add a module to does have one?
18:30:54 <acozine> usually a CLA says "I, the contributor, understand that neither I nor my employer has any right to claim this contribution, and that the repository's owner does have that right"
18:31:05 <samccann> thanks
18:31:10 <Leo[m]> samccann: yes, that's why I wouldn't rush it. going one way or the other should be researched a bit imo
18:31:17 <acozine> samccann: they'd notice if they try to contribute
18:31:31 <acozine> the contribution won't be accepted unless they sign the CLA
18:31:51 <samccann> just not sure how frequently an open source contributor hits that CLA in their day to day so to speak. Is it rare, or everyone allows it but potentially Ansible so to speak
18:32:09 <gotmax23> https://paste.sr.ht/~gotmax23/00df144f8b0d12f3017e575e53ad0222fa5239d6
18:32:11 <samccann> acozine: thanks that helps
18:32:32 <acozine> I don't immediately know of other collections that have one. But again, you wouldn't necessarily know until/unless you contribute
18:33:09 <samccann> Thanks gotmax23 !  so we have a small handful that include CLAs already
18:33:24 <gotmax23> I did some naiive grepping. it looks like the purestorage collections and the netapp_eseries collections do.
18:34:05 <samccann> So I agree w/ Leo, we should probably put more thoughts/ideas in the discussion
18:34:51 <gotmax23> To note, here is what's written in Red Hat's open source participation guidelines
18:34:57 <gotmax23> > Red Hat-led projects do not use contributor license agreements (CLAs), copyright assignment, or other formal contributor agreements, apart from rare exceptions specifically approved by Red Hat Legal. We have learned from experience that these mechanisms can be a significant barrier to building communities around the projects we sponsor. However, we encourage all Red Hat-led projects to use the DCO.
18:35:12 <samccann> as for what it means for us to say no CLAs.. and to what extent we might 'harm' the user community, vs allow CLAs and... the contributor decides at PR time whether to sign it
18:35:42 <gotmax23> (Fedora is one of those "rare exceptions" they're trying to get rid of, but the Fedora CLA doesn't actually require copyright assignment)
18:37:29 <gotmax23> I don't think contributors should be asked to sign away their rights or have other hindrances to contribution. If collections want to have these, than they can always be on Galaxy but not in the package.
18:37:30 <acozine> I think there are two major pitfalls with CLAs. The first one is that signing one puts a lot of people off - they think "I don't really know what this says, I'm not going to sign it right now" and they never get back around to it. So we lose contributors and contributions.
18:38:15 <acozine> The second pitfall is that there can be hard feelings if someone else starts making money off some code that a community contributor wrote
18:38:23 <gotmax23> Even if the contributor is fine with signing the CLA, they may still have to get approval from their employers's legal department which is definitely a burden to contribution
18:38:42 <acozine> hard feelings, maybe lawsuits, American society being what it is
18:38:49 <gotmax23> acozine: 100%
18:39:07 <Leo[m]> samccann: While we investigate the issue, I like the approach of leaving it to the contributor for those collections that have/need it, but mentioning the guidelines quoted by gotmax23 somewhere in the inclusion rules for community namespace for example.
18:39:52 <cyberpear> there's also the problem if someone has personally signed the CLA but then works something on the project on company time but the company hasn't signed
18:39:57 <acozine> I worked on a completely-open-source project for library software that required a CLA and a lot of folks just couldn't contribute because they couldn't get approval to sign the CLA, which covered the person's employer.
18:40:15 <cyberpear> (that's part of why I won't sign one personally because I don't want to be "able" to accidentally make that mistake)
18:40:30 <gotmax23> yeah, if we can't get consensus to ban them completely for new collections, we could start with something like collections in the community namespace MUST NOT have these, but other collections are strongly encouraged not to use them
18:40:54 <acozine> so I guess my take is a Free-and-open-source project that has a CLA is not quite so free-and-open-source
18:41:12 <Leo[m]> acozine: and that would harm the collection itself in the end, wouldn't it? I mean, the best reason to not require it, is for them to understand they wont get (as many) contributions if they do.
18:41:18 <acozine> Leo: yes
18:41:23 <samccann> Sounds like we should ping more people to participate in that discussion before we vote
18:41:35 <gotmax23> yeah, it has multiple downsides. it adds hindrance, takes away community ownership, and may bar someone from contributing all together whether they have legal or moral objections
18:41:47 <gotmax23> samccann: SGTM
18:41:57 <acozine> and what harms collections harm the overall project too
18:42:08 <acozine> s/harm/harms/
18:42:27 <samccann> #info please add your comments to the issue on whether or not to band CLAs - https://github.com/ansible-community/community-topics/issues/221
18:43:17 <samccann> yeah I'm leaning to it could have a negative 'mark' on the whole Ansible community/ecosystem, even thought it's only one minor part of it.
18:43:27 <gotmax23> might want to #undo that and fix that band typo :)
18:43:29 <samccann> #undo
18:43:29 <zodbot> Removing item from minutes: INFO by samccann at 18:42:27 : please add your comments to the issue on whether or not to band CLAs - https://github.com/ansible-community/community-topics/issues/221
18:44:08 <samccann> #info addo your comments to the issue on whether or not to ban CLAs - https://github.com/ansible-community/community-topics/issues/221
18:44:21 <samccann> ok I have one quick topic and then we can open the floor
18:44:30 <samccann> #topic Ansible roadmaps
18:44:42 <samccann> #info - created a project roadmap so we track alpha/beta etc releases - https://github.com/orgs/ansible-community/projects/7/views/1
18:44:45 <samccann> #info documentation roadmap will remain the cannonical source. Project board it just to help us keep on track.
18:44:56 <samccann> Is this useful? Are there other items we want to add?
18:45:00 <remindbot[m]> @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting
18:45:36 <gotmax23> I think having issues to track release dates are a great idea. Thanks for doing that!
18:45:59 <samccann> cool. if there are more issues to track, go ahead and add them or ping me if you need help
18:46:00 <samccann> #info Ansible 9 roadmap work needs to start. discussion: https://github.com/ansible-community/community-topics/issues/222
18:46:12 <samccann> ... in the hopes of getting a roadmap out on time :-)
18:46:26 <samccann> #topic Open Floor
18:46:33 <samccann> Anyone have a topic to bring up?
18:47:24 <cybette_> I should add those dates as reminders in bullhorn as well
18:47:25 * gotmax23 looks
18:47:55 <samccann> #info community website WG is actively creating a prototype! If interested in web design/UX - head over to https://matrix.to/#/#website:ansible.com to see what's happening
18:48:14 <samccann> that's about alls I got
18:48:25 <samccann> anything else lingering or should I end the meeting?
18:48:40 <gotmax23> #info ansible 8.0.0a1 was released today: https://groups.google.com/g/ansible-announce/c/MPRZTafDerA
18:48:51 <samccann> woot!!
18:49:00 <samccann> Thanks anwesha !!
18:49:06 <gotmax23> I had https://github.com/ansible-community/community-topics/issues/218
18:49:52 <gotmax23> I don't have anything to update about that and still need to put together a draft PR but wanted to remind y'all
18:50:21 <acozine> gotmax23: it seems like a good idea to me, keep everybody up-to-date
18:50:28 <samccann> #info take a look at https://github.com/ansible-community/community-topics/issues/218
18:50:30 <acozine> as long as the release managers are okay with it
18:50:45 <gotmax23> yeah, my one question was where this should be documented
18:50:52 <acozine> heh
18:51:00 <acozine> on the new website . . . ?
18:51:11 <gotmax23> I don't think the current ansible release process is documented anywhere
18:51:31 <gotmax23> IIRC, someone said there's some internal documentation
18:51:57 <acozine> we could put it in the same spot as the documentation about including collections
18:51:58 <gotmax23> but one of the requests was that I document this new process and how to handle pinning collections, etc.
18:51:58 <samccann> I only skimmed the issue - I thought this is a step the collection maintainers do?
18:52:31 <acozine> samccann: yeah, that's why I was thinking document it next to collection inclusion
18:52:33 <acozine> same audience
18:52:45 <samccann> we have collection developer docs scattered in a few places. andersson007_ might know of the top of his head when he's back tomorrow
18:52:50 <gotmax23> I was referring to the ansible community package release process
18:52:52 <acozine> only I can't remember where that is
18:53:06 <Leo[m]> acozine: this? https://github.com/ansible-collections/ansible-inclusion
18:53:07 <samccann> yeah we don't document the community package release process anywhere yet afaik
18:53:12 <gotmax23> specifically this comment: https://github.com/ansible-community/community-topics/issues/218#issuecomment-1489105440
18:53:47 <acozine> samccann: I meant, I can't remember where we put the rules for removing collections, adding collections, etc.
18:54:04 <gotmax23> I think that's in ansible-build-data
18:54:12 <samccann> yeah, that's scattered too
18:54:25 <acozine> Leo: thanks, that's the right content
18:54:36 <acozine> somehow I remembered it as a markdown page
18:54:52 <acozine> anyway, gotmax23 you could do a docs PR against that page to start with
18:55:01 <acozine> if folks think there's a better place, they'll chime in
18:55:09 <gotmax23> but I'm not sure there's anything about releasing a new version (i.e. filing the PRs, running the release process, uploading the package to PyPI, announcing on the mailing list, etc.)
18:55:18 <gotmax23> s/process/playbook/
18:57:09 <acozine> ah-HAH! What about https://docs.ansible.com/ansible/latest/community/collection_contributors/collection_requirements.html#collections-requirements?
18:57:37 <acozine> we could add a header for Tagging there
18:57:50 * gotmax23 worries he's not being clear
18:57:53 <acozine> gotmax23: I agree we should add documentation about hte release process
18:58:13 <acozine> but I think we'd need more information before we could do that
18:58:26 <acozine> at least, i don't know enough about how it works to document it
18:58:36 <gotmax23> :nod:
18:59:00 <acozine> I do hear you - there is no documentation about uploading to PyPI or any other parts of the release process
18:59:14 <acozine> but if we let that be a blocker for the Tagging docs, we might have to wait a while
18:59:44 <samccann> yeah it's a known issue. Parts of it I think were only 'doable' by a RH'er. I think that's changed and there's probable internal docs somewhere we can work to get external
19:00:12 <gotmax23> I'm mainly asking, because I was asked to document the work that the release managers would need to do as part of this proposal, but I don't know where to do that, as the rest of the process isn't documented
19:00:46 <cybette_> I think it could be detailed in this issue, but definitely requires more input and info https://github.com/ansible-community/community-team/issues/160
19:01:29 <gotmax23> cool, thanks.
19:01:29 <samccann> thanks cybette !
19:01:35 <samccann> anything else?
19:02:27 <gotmax23> In the meantime, I can write up documentation about this specific process in my draft ansible-build-data PR and then it can be reorganized later
19:02:36 <samccann> cool thanks
19:02:47 <gotmax23> it can just be a separate section of the README
19:03:41 <acozine> thanks gotmax23 ! any documentation expansion is a good thing, even if it's partial and not in the perfect location
19:03:51 <gotmax23> yeah
19:04:05 <gotmax23> does anybody have any other topics or something else to add to this?
19:04:17 <acozine> I put two comments in the issue
19:04:27 <acozine> otherwise I'll forget by next week
19:05:02 <samccann> ok sounds like a good stopping point
19:05:30 <acozine> thanks for running the meeting samccann
19:05:35 <samccann> #endmeeting