15:00:58 #startmeeting Ansible Documentation Working Group aka DaWGs 15:00:58 Meeting started Tue Aug 8 15:00:58 2023 UTC. 15:00:58 This meeting is logged and archived in a public location. 15:00:58 The chair is samccann_irc. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 15:00:58 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:58 The meeting name has been set to 'ansible_documentation_working_group_aka_dawgs' 15:01:24 #info for the time being, these meetings must be run from IRC. Matrix users can participate but 15:01:25 can't use any of the typical meeting commands. 15:01:28 06TVo> Hey there, I'll be 🪰 on the wall again, I have another community office hour call at the same time. 15:01:40 12samccann> @room Meeting time! Who is here to talk the docs? 15:01:56 02Don Naro> o/ 15:02:00 02Don Naro> hi 15:02:06 12samccann> welcome Tvo! Can't chair ya but .. 15:02:12 12samccann> #sofa Tvo, Don 15:02:21 o/ 15:02:24 #chair felixfontein 15:02:24 Current chairs: felixfontein samccann_irc 15:02:47 14sutapa_bhattacharjee> O/ 👋 15:02:48 hi sofa :D 15:02:56 hehe welcome Sutapa! 15:03:22 As folks come in - a bit of explanation. Our bridge between irc and matrix is down. We have a relaybot that allows us to see each other's messages 15:03:35 but the actual meeting has to be run from IRC.. thus my extended nick etc 15:04:12 Raise your ascii hand (o/) to say hi or any other way you want to let us know you are here. And Welcome to any new folks! 15:04:12 To any newcomers - again, welcome. We chair all attendees as a way of recognizing your time spent here. And it opens it up for people to add to the meeting minutes with commands like #info or #link (to add a link) 15:04:12 General run of the meeting - We go over action items, give docs updates.. maybe have a topic or two, and go over doctooling updates (all the fun stuff behind the scenes that get us docs.ansible.com!) 15:04:24 official agenda at https://github.com/ansible/community/issues/678#issuecomment-1660638342 15:04:41 #topic Documentation Updates 15:04:50 #info we have 40 open issues and 16 open PRs 15:05:00 and backport bot is working great! 15:05:09 best invention since sliced bread 15:05:22 definitely! 15:05:32 #info awx operator docs nearly there! just need to set up readthedocs and we can publish! 15:06:05 oh watching this on the matrix side and every hashtag use here ends up a heading (aka large font) on matrix. fun times 15:06:19 #topic Build issues w new dependencies files 15:06:40 Currently, the jenkins builds will only work with the updated dependencies work from https://github.com/ansible/ansible-documentation/pull/60 15:06:49 Our options seem to be - make the jenkins logic more complex so it still works down to at least stable-2.13 15:07:01 or manually 'backport' that new dependencies option down to stable-2.13. 15:07:12 I think manually backport is the better approach 15:07:31 the 'hard' part of it is re-creating the pip-compile result with a similar Python version 15:07:32 That still leaves us unable to publish EOL docs. Historically, we've had to fidget with them and thus needed the ability to publish. 15:07:57 02Don Naro> I'm currently wrangling with those dependencies now and will hopefully be able to fix it soon enough 15:08:07 I wanted to try this out, but haven't gotten around to do it yet... not sure whether I will before the weekend, so if anyone else wants to try, go ahead :) 15:08:22 we need the ability to publish to latest working at least (aka stable-2.15) 15:08:53 I bakported #60 PR (The new pinning) and am running a test jenkins build to see if it works 15:09:07 s/bakported/backported/ 15:09:40 do you already have a PR for that? 15:09:56 me? the stable-2.15 I think merged... lemme check 15:10:09 02Don Naro> yeah stable-2.15 backport was merged 15:10:39 ok phew. I'll letcha know when the jenkins build completes. That's my priority - getting latest publish back 15:11:09 anything else to bring up on pinning/not pinning requirements? 15:11:17 ah, I thought you meant a backport for 2.14 or 2.13 15:11:45 nah. Don has more work going there so a manual backport to those two might as well wait for him to finish 15:12:10 unless I'm missing something and someone should go ahead and do the backports for #60 now? 15:12:52 02Don Naro> samcann: and Felix Fontein https://github.com/ansible/ansible-documentation/pull/255 - if you could take a look at this one, I'd appreciate it 15:13:23 there will be new ansible-core 2.13 and 2.14 releases next Monday or Tuesday 15:13:45 so being able to build their docs would be helpful :) 15:14:36 ooch yeah 15:14:38 which Python version(s) is/are used for the main docs build? 15:14:55 main as in devel/latest? 15:15:42 no, main as in "the official docs build for docs.ansible.com" 15:15:52 02Don Naro> 3.11 15:15:55 (main wasn't the best word for it ;) ) 15:16:08 yeah so docs.ansible.com has two 'main' builds - devel and latest and yeah both use 3.11+ 15:16:32 ansible-core 2.13 does not support Python 3.11 15:16:45 I think the jenkins build has logic for that 15:17:13 lemme check unless don knows off the top of his head. The core branches are a separate jenkins build from the Ansible package jenkins build 15:17:36 02Don Naro> bah, I don't remember off the top of my head either 15:17:45 I guess we should use the Python version used by Jenkins for every stable branch to compile the requirements for that stable branch 15:18:24 ok I don;t have the jenkins repo handy, but looking at the builds that failed, they are using 3.11 as well 15:18:57 the jenkins list of older builds only show me the last 3, which were all after #60 where I found the builds broken for all but devel 15:19:59 Yeah it's all gotten beyond my ability to track. Don would be the one to help with what's required for backports to get stable-2.13, 2.14 working before Monday 15:20:44 02Don Naro> it's a bit of a tangled mess of dependencies but I'm working on it 15:20:52 Don - can you resolve your new PR tomorrow, and then work on the older branches, or should we backport #60 manually to the older branches and get publishing working on them first... then finish your new PR after next week's releases? 15:21:32 02Don Naro> I think we can do the backports independently of that other PR I created 15:21:54 sounds good to me 15:22:10 ok who can do them? cuz this has gotten way over my head so I'm not help 15:22:18 s/not/no/ 15:22:25 * samccann_irc types worse in irc ;-) 15:23:14 I can work on them, but I need to know which Python is used by Jenkins to build the docs for these stable branches 15:23:42 02Don Naro> can you see my messages on IRC? 15:23:48 yes 15:25:07 02Don Naro> ok. I've been looking at the backports and am making them a priority. I hope to have PRs for the stable branches soon, like tomorrow. 15:25:34 02Don Naro> I'd def appreciate any help though Felix Fontein 15:25:38 ok, sounds good to me 15:25:46 02Don Naro> cool 15:25:58 great thanks! 15:26:01 02Don Naro> yeah, Jenkins uses 3.11 for all the branches 15:26:20 02Don Naro> we really need to open those jobs up 15:26:45 02Don Naro> it also feels like a good time to ask about possibly changing the versioning for docs 15:27:24 yep definitely need to open them up. 15:27:33 did you want me to start a new topic? 15:27:39 02Don Naro> let's get the requirements thing sorted first but do we still want to try and move to a devel / latest only schema in the repo? 15:27:41 (aka versioning for docs?) 15:27:56 not sure what you mean? 15:28:02 what exactly do you mean with 'versioning for docs'? 15:28:16 (sorry, typing slowly today ;) ) 15:29:10 02Don Naro> mostly the fact that the docs repo still follows the core branch names. `latest` package docs url builds off `stable-2.15` etc. 15:29:23 ok lemme start another topic 15:29:44 #topic ansible-documentation release tagging 15:29:54 which kind of is necessary, as some parts of the docs are version specific 15:30:02 #info https://github.com/ansible/ansible-documentation/issues/66 15:30:26 is related to this, tho not identical to what Don's talking about i think? 15:31:41 I'm not entirely sure 15:32:03 also latest means different things, depending on whether you are talking about the ansible-core or the ansible docs (at least for some time of the year) 15:32:43 12maxamillion:one.ems.host> whoa ... relaybot looks ... weird 15:32:53 ok I might have mixed apples w oranges on that issue 15:33:03 maxamillion: indeed! 15:33:19 hey maxamillion - yeah it gets messy but better than the compete isolation we had a couple of days ago 15:33:55 the big thing is if you run meetings w/ zodbot, you end up having to run them from irc 15:34:52 12maxamillion:one.ems.host> oh yeah, it's certainly better than complete isolation :) 15:35:02 12maxamillion:one.ems.host> I'll get used to it ;) 15:35:25 12maxamillion:one.ems.host> oh that's a bummer 15:35:43 12maxamillion:one.ems.host> hadn't thought about that 15:36:03 12maxamillion:one.ems.host> time for a new bot? 15:36:15 12samccann> Don Naro: sorry we digressed. What did you want to bring up about latest/devel 'branches' in the docs? 15:36:23 12maxamillion:one.ems.host> a matrix native one? 15:36:40 maxamillion: right now a matrix native bot won't really help, as we'd still need the same crutch for IRC folks 15:36:54 the main difference is that it looks better for the matrix folks and worse for the IRC folks ;) 15:37:15 12samccann> maxamillion: yeah there's rumor'd work happening about a meeting bot or something for matrix, but we still have the irc vs matrix communications problems 15:38:43 02Don Naro> samccann: I guess we need to keep the branch naming scheme that we have now if some bits need versioning. I was just wondering the other day if we might try and change that and forgot where things are at, or if there's any need/desire to change what we're doing. 15:39:51 12maxamillion:one.ems.host> ahhh ok 15:39:58 ok thanks 15:40:09 #topic Preferred terminology 15:40:09 12maxamillion:one.ems.host> interop fun continues! 15:40:19 12maxamillion:one.ems.host> alright, sorry ... I'll stop derailing from the meeting 15:40:24 Don: yeah, I guess we first have to figure out how to separate the version-specific things from the version-agnostic things... which will be hard by itself, and might need another repo... 15:40:48 #info adding project terms to style guide - https://github.com/ansible/ansible-documentation/issues/152. 15:40:57 (or maybe the version specific things could go back to ansible/ansible, so they are versioned together with the code) 15:41:15 Seems we are coming to consensus there wrt using capitals etc. I'll update the hackmd one more time and then open a PR to add this to our style guide 15:41:44 felixfontein: docs can't go back to ansible/ansible. 15:42:28 #topic Github metrics 15:42:29 that's a policy decision, not a technical decision, so it could be changed in theory ;) just mentioning it, since it might be nicer to have two repos instead of three 15:42:44 heh 15:42:51 #info do we want to turn on metrics ? see https://github.blog/2023-07-19-metrics-for-issues-pull-requests-and-discussions/ 15:43:04 someone pointed that link out so figured I'd see if there's any interest 15:44:38 I think that GHA is only using public information that's already there anyway - is that correct? 15:44:53 if that's the case I don't see a reason (from a privacy point of view) against enabling it 15:45:30 it sounds to me like this is similar to what Greg has been doing in the past for other repos 15:45:42 yeah just figured this one we can set up 15:46:00 12samccann> but worth pinging Gwmngilfen to see if he has an opinion 15:46:06 :+1: 15:46:23 (yay, that was successfully converted to an emoticon by the 'bridge' :D ) 15:46:45 hehe yep sometimes it really is just the little things :-) 15:46:51 #topic Open Floor 15:47:05 We have a few minutes left. Anyone have a docs-related topic to discuss? 15:48:04 08Gwmngilfen> just to check, this is stats on the new docs repo? 15:48:22 yeah this is the new github thing that does stats 15:48:32 https://github.blog/2023-07-19-metrics-for-issues-pull-requests-and-discussions/ 15:48:43 we're thinking of enabling it on the docs repo for fun 15:48:54 08Gwmngilfen> i hadn't seen it, too busy. let me go read what it does and I can give an opinion. but in any case I'd be happy to do one of my reports for the repo too 15:49:25 ok let us know what you think when you get a chance 15:49:33 I'm not sure whether the docs repo is a good repo for trying this out, since it's history isn't that long 15:49:34 08Gwmngilfen> ugh average time to close. that is *always* a lie if you just use a mean (which they might well be) 15:49:41 :-) 15:49:52 anything else to talk about before we end the meeting today? 15:50:47 ok sounds like that's a wrap! 15:50:51 #endmeeting