14:03:55 <cybette> #startmeeting AnsibleFest Austin: Contributors Summit - Documentation
14:03:55 <zodbot> Meeting started Thu Oct  4 14:03:55 2018 UTC.
14:03:55 <zodbot> This meeting is logged and archived in a public location.
14:03:55 <zodbot> The chair is cybette. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:03:55 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
14:03:55 <zodbot> The meeting name has been set to 'ansiblefest_austin:_contributors_summit_-_documentation'
14:04:06 <dag> Aha movement ! :)
14:04:16 <cybette> #info etherpad link https://etherpad.openstack.org/p/ansible-summit-october-2018-documentation
14:04:28 <cybette> #info bluejeans link https://bluejeans.com/7025862841/
14:05:03 <felixfontein> is there already sound?
14:06:20 <cybette> felixfontein: can you hear us?
14:07:16 <defionscode> gundalow: the rooms are split up now? main room is docs, and the others?
14:07:41 <bcoca> tried to join x2 bluejeans sessions and now im blocked from any/all of em
14:07:59 <gundalow> 1 is docs
14:07:59 <gundalow> 2 is windows
14:07:59 <gundalow> 3 is galaxy
14:08:33 <defionscode> ah ok, thanks! AWS is later or was is pulled off?
14:09:09 <dag> bcoca: I have 2 open, but one hasn't started, might still fail :-)
14:09:59 <cybette> defionscode: AWS is later at 13:00 Austin time
14:10:20 <defionscode> cybette: ok cool
14:11:10 <felixfontein> audio is hard to understand
14:11:15 <felixfontein> bits seem missing
14:11:31 <dag> felixfontein: I think this is background chatter ?
14:11:40 <felixfontein> ok, that would be fine :)
14:12:52 <ryansb> AWS is on today
14:12:57 <dag> felixfontein: I hope it is :)
14:13:11 <samccann> dag: same here
14:13:18 <felixfontein> oooh, ansible 2.7 has just been released
14:13:39 <felixfontein> yes
14:14:31 <felixfontein> I don't have a mic
14:15:07 <abadger1999> Okay, gundalow is going to help get bluejeans and the room's microphones setup
14:15:43 <samccann> \o/
14:16:11 <geerlingguy> I am back in St. Louis... and I have lost context—can someone link to the etherpad for today's meetings?
14:16:25 <felixfontein> https://etherpad.openstack.org/p/ansible-summit-october-2018-documentation
14:16:26 <geerlingguy> would like to try to join one or two when I'm not stuck in a meeting
14:16:27 <geerlingguy> thanks!
14:16:49 <felixfontein> cut off after 2018 to get the general one
14:17:16 <felixfontein> yes
14:17:20 <felixfontein> whoever is speaking
14:17:43 <felixfontein> hearing you too
14:18:57 <cybette> bluejeans link again: https://bluejeans.com/7025862841/
14:19:49 <geerlingguy> I'll be able to join soon, still getting going after a late night of travel
14:21:54 <geerlingguy> Is documentation the only breakout happening currently? The Galaxy session bluejeans link was invalid
14:22:13 * sdoran was wondering the same thing
14:22:17 <bcoca> galaxy is working for me
14:22:25 <bcoca> and they are presenting
14:22:30 <geerlingguy> https://bluejeans.com/748091404391 :/
14:22:31 <bcoca> https://bluejeans.com/7480904391/
14:22:40 <geerlingguy> ah, someone stuck a 14 in there
14:22:44 <bcoca> i did have a lot of trouble
14:22:48 <dag> audio is still a bit soft (compared to other noises),
14:23:00 <felixfontein> dag+
14:23:14 <felixfontein> sounds like someone is installing something audio-related
14:23:23 <felixfontein> very close to the mic at least
14:23:23 <dag> can everyone else please shut their microphone
14:23:34 <sdoran> That's better.
14:23:39 <dag> (ah everyone has)
14:24:01 <felixfontein> is that gundalow?
14:24:11 <sdoran> That's alikins
14:24:17 <bcoca> Adrian
14:24:18 <felixfontein> ah
14:24:37 <dag> so alikins audio is very good, but I can't hardly hear Carol
14:24:41 <bcoca> less british accent
14:25:00 <bcoca> i 'unjoined' docs, could not hear anything clearly
14:25:17 <dag> bcoca: they fixed it
14:25:21 <dag> its fine now
14:25:39 <felixfontein> audio is fine now
14:25:43 <felixfontein> <-- same timezone as dag
14:25:48 <abadger1999> From room: Desire to be able to jump on irc and get involved in working on something.
14:25:54 <felixfontein> did someone else say something?
14:25:58 <cybette> yes
14:26:01 <felixfontein> then no
14:26:12 <felixfontein> at least in the last 30 seconds or so
14:26:15 <felixfontein> testingtesting123
14:26:16 <cybette> chris is talking
14:26:29 <abadger1999> Question about which timezones people are in.
14:26:45 <felixfontein> CEST
14:26:52 <abadger1999> (Audio is now working so I'll stop transcribing)
14:27:00 <felixfontein> abadger1999: thanks!
14:27:02 <sdoran> US East (UTC -0400)
14:27:11 <samccann> US east
14:27:28 <felixfontein> I'm interested in module docs
14:29:16 <felixfontein> for me it's not (frustrating thing repeating myself), but it's easy to forget to change default values/... in both places, so having some things (default, type, choices) in only one place is kind of neat for me
14:29:33 * geerlingguy will be back in a while
14:30:57 <dag> So we do want to parse the Windows arg_spec as well, but this is being reworked now anyway, so we are not looking at that at this time
14:30:57 <abadger1999> This is dag's PR: https://github.com/ansible/ansible/pull/45805
14:31:04 <ChrisShort> ty
14:31:15 <dag> as soon as we have a well-defined arg_spec (which is almost done) we can look at that
14:31:32 <dag> the question remains do we want to pull in pwsh for generating docs ?
14:32:19 <felixfontein> duplication = sanity check is a very good point
14:32:28 <dag> but even if that answer is no, there's still value in having the argument spec being used for documentation
14:32:28 <sdoran> I'm working on that for 2.8 (removing arg_spec from AnsibleModule).
14:32:57 <dag> felixfontein: I don't agree, that's like saying, let's implement every module twice to see if there's no bug introduced
14:33:18 <dag> felixfontein: while certainly valuable, that is not practical
14:33:34 <felixfontein> dag: I don't say we shouldn'd do it because of that, but that it is a good point I haven't thought about so far :)
14:35:51 <defionscode> dag: I mean whats an alternative? make argspec super verbose? ¯\_(ツ)_/¯
14:36:16 <dag> oh wait
14:36:24 <dag> the argument_spec is not the only thing used
14:36:41 <dag> the documentation is still be used for descriptions
14:36:45 <dag> being
14:38:15 <defionscode> oh i see, dag I hadnt seen that PR, my initial reaction is "i like"
14:39:30 <dag> and the documentation still wins over the argument spec
14:40:48 <cybette__> #link https://github.com/ansible/ansible/pull/45805
14:42:43 <felixfontein> do I understand it correctly that for Windows modules, there's usually a .py file which is only for documentation, and the real module is the .ps1 file? (I've never looked at a Windows module before)
14:42:55 <defionscode> felixfontein: that's correct
14:44:01 <defionscode> and there is no argspec-like interface either afaik
14:44:19 <felixfontein> yep, it looks like (at least in the 1-2 examples I looked at ;) )
14:44:30 <felixfontein> so nothing structured yet
14:45:20 <felixfontein> there's one file moved to lib/ansible/utils/_module_args.py in the PR
14:45:40 <bcoca> there is a 'powershell argspec' but it works very differently
14:45:57 <bcoca> but i do think it is close to 'feature parity' with the python one
14:47:49 <abadger1999> I think I like the high level of this feature.  But _module_args.py makes use of mock.  So we wouldn't want anything to need that at runtime.
14:48:00 <defionscode> logic wise maybe we'd have to do an open() on the ps1 and parse out the ps argspec portion?
14:48:08 <cybette__> #link Guide to contributing to docs (WIP) https://github.com/ansible/ansible/pull/46481
14:48:46 <dag> defionscode: This is the Windows arg_spec implementation: https://github.com/ansible/ansible/pull/44705
14:48:47 <bcoca> just realized, the argspec conversation in galaxy is not this one .. this is one in docs land
14:49:05 <defionscode> ah
14:49:09 <bcoca> https://github.com/ansible/ansible/compare/devel...alikins:argument_spec
14:49:32 <bcoca> ^ adrian's branch trying to put args spec checks outside of AnsibleModule for better reuse
14:49:40 <felixfontein> dag: could it be that something got lost?
14:50:42 <dag> felixfontein: might be, fact is that I removed the example modules so maybe...
14:51:08 <felixfontein> too bad there isn't an older version of the PR still lying around... :)
14:52:37 <defionscode> i agree re: explaining what happens
14:52:53 <felixfontein> the only page I can see is the PR right now
14:53:04 <felixfontein> ah, now there's a different one
14:53:46 <felixfontein> *nod*
14:56:09 <gundalow> For those attending remotely, are you managing to follow along OK? Please do keep your ideas coming.
14:56:28 <dag> bcoca: That'll be very useful
14:56:52 <cybette__> #link Ansible Style Guide https://docs.ansible.com/ansible/latest/dev_guide/style_guide/index.html
14:57:16 <bcoca> part of the 'role spec validation' uses that code, but it currently 'fakes a module' , factoring out argspec for use in other systems outisde AnsibleModule seems to benefit docs also
14:57:19 * abadger1999 notes that he is trying to record the ideas from the meeting into the etherpad
14:57:41 <dag> bcoca: I would like to validate the EXAMPLES section by the arg_spec too
14:57:52 <dag> abadger1999: very much appreciated !
14:57:58 <bcoca> another need role spec checking has is 'descriptions/labels' which if added to argspec might simplify migrating docs
14:58:00 <cybette__> abadger1999: great job, thank you <3
14:58:25 <felixfontein> I did not either
15:00:07 <felixfontein> a mail to ansible-devel mailinglist would be helpful to find about such stuff as well
15:00:20 <geerlingguy> back soon, had to drop again
15:01:44 <dag> felixfontein: or the Contributor's Newsletter ! :-)
15:01:56 <Viper233> https://docs.ansible.com/ansible/latest/dev_guide/developing_modules_documenting.html#documentation-block I don't see the style guide linked  here
15:02:03 <felixfontein> dag: yeah, in whatever form it appears :)
15:02:14 <dag> felixfontein: printed on glossy paper !
15:02:31 <felixfontein> dag: there was this discussion in some ansible/community issue about how to propagate info on such things which affect contributers
15:03:00 <dag> felixfontein: I know, have been asking for this since London, to build a contributor's community
15:04:56 <felixfontein> let's pester gundalow about that in the contributer experience session afterwards
15:06:20 <gundalow> :)
15:07:57 <felixfontein> oh wait, there is no more contributer experience session today, right?
15:08:02 <felixfontein> I think I looked at the wrong day
15:08:34 <felixfontein> anyway, that's something for your bucket list gundalow ;)
15:10:32 <felixfontein> most people aren't committers
15:10:40 <felixfontein> so for most people, that's not a problem :)
15:11:19 <gundalow> felixfontein: we can sync up when I'm back
15:11:28 <felixfontein> gundalow+
15:12:08 <felixfontein> gundalow: I wrote myself a note
15:16:03 <Viper233> https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-apigateway-domainname.html example of edit on github page
15:18:13 <felixfontein> adding to what dag says: also when switching between documentations, renamed modules make problems
15:18:36 <felixfontein> (maybe if a module has an alias, a dummy file should be generated at the alias place to redirect to the new location?)
15:18:45 <felixfontein> (assuming module renames keep an alias for a bit of time)
15:18:55 <bcoca> i thought we fixed that a couple of months ago
15:19:24 <bcoca> alias is one function, deprecation uses it, but having an alias does not automatically mean it is deprecated
15:19:56 <bcoca> shell: rm -rf /
15:20:04 <dag> Viper233: I don't see an edit on github button ?
15:20:10 <felixfontein> it doesn't seem to work: letsencrypt is an alias of acme_certificate (a deprecated one), but https://docs.ansible.com/ansible/latest/modules/letsencrypt_module.html does not exist while https://docs.ansible.com/ansible/latest/modules/acme_certificate_module.html works
15:20:11 <dag> ah, the logo
15:20:40 <dag> Viper233: I prefer adding the text there, wouldn't be surprised that people miss out on it
15:21:12 <dag> felixfontein: We have restructured so many things that it would leave behind a lot of ghost-files :-(
15:21:38 <felixfontein> dag: if they're just there while the alias is actively mentioned in the current version, that's fine for me
15:21:39 <dag> felixfontein: however this is something that over time no longer poses a problem
15:21:48 <dag> felixfontein: unless we are restructuring a lot :)
15:21:48 <Viper233> dag: just the logo. It's still just an 'Edit on github' process. This seems to work is this community. Text/logo doesn't matter, the fork/commit/PR user experience is the same
15:23:00 <felixfontein> dag: but at least for aliases (which are not deprecated), shouldn't it be possible to put the alias into the URL to get to the documentation?
15:23:11 <dag> Viper233: I disagree, the logo alone does not give sufficient clues, the text provides a mental note
15:23:32 <bcoca> big red flashing banner
15:23:50 <Viper233> <blink>
15:23:50 <dag> felixfontein: yes, for aliases a solution could be built
15:23:53 <dag> bcoca: :)
15:23:57 <felixfontein> bcoca: <marquee>?
15:24:28 <bcoca> <marquee><blink>...
15:24:40 <felixfontein> yeah
15:25:20 <Viper233> dag: I think you are right. I was looking for text. I already knew that you could edit the documentation on github, it took me a couple of minutes to figure out that it was the logo. I thought initially it was a SSO with my github account
15:25:28 <cybette__> #info Monthly "fixit" topics for the WG - let's pick 3 for Oct/Nov/Dec - vote by adding +1 next to the topic https://etherpad.openstack.org/p/ansible-summit-october-2018-documentation
15:25:56 <felixfontein> something about docs: I've sometimes seen people using markdown styling in module docs, mostly single backticks. Maybe there could be some sanity check telling people not to do that (and to use C(...) or I(...) instead)? :)
15:26:11 <dag> We are targeting "Drive by contributors" (actually users) mostly with this
15:26:40 <dag> those that are sufficiently frustrated that they want to avoid the next purpose going through the same ordeal
15:27:11 <dag> felixfontein: Yes, tests for that would be a good idea !
15:27:20 <dag> felixfontein: I see them do **bold** too :)
15:28:08 <dag> So I lost some of the audio the past 5 minutes as the kids returned from school :-(
15:29:39 <dag> Yes, don't forget EMEA :-P
15:30:05 <cybette__> break for 15 min, back at 10:45 CDT
15:30:17 <dag> Thanks !
15:30:17 <felixfontein> 17:45 CEST :)
15:31:01 <dag> gundalow: take a t-shirt back for me !
15:31:06 <dag> gundalow: Large will do :-)
15:31:25 <felixfontein> actually, what about having some more "helpers" like C(...), I(...), U(...), L(...,...)?
15:31:44 <felixfontein> and how about making them work nicely in ansible-doc as well (at least L is pretty ugly there)?
15:32:26 <gundalow> Dag the t-shirts ran out on the first day. I did t even get one :(
15:32:59 <abadger1999> felixfontein: If you have ideas, that would be a good area to improve.
15:33:34 <dag> gundalow: bummer
15:33:59 <dag> felixfontein: there's a feature request open to redo ansible-doc, because I hate the output
15:34:48 <dag> https://github.com/ansible/ansible/issues/46011
15:34:48 <abadger1999> We've talked about things like how to make links better but it wasn't obvious to us how to format it better
15:34:59 <bcoca> current output is very basic
15:35:01 <felixfontein> dag: I think the output of ansible-doc is not *that* bad ;) it's mainly L(...,...) which is ugly
15:35:14 <felixfontein> abadger1999: how about "`link name' (URL)"?
15:35:29 <dag> abadger1999: We did make the link generation better, it includes the URL now (and in some terminals you can click now)
15:35:44 <dag> felixfontein: Isn't that L(name,url) ?
15:36:02 <felixfontein> dag: either that or L(url,name). I always have to check :)
15:36:52 <dag> felixfontein: The changes are in here: https://github.com/ansible/ansible/pull/45949
15:37:19 <dag> but this was for the See also stuff, we should do the same for L()
15:37:58 <bcoca> +1, just to bikeshed: 'related'
15:38:13 <felixfontein> lol
15:38:21 <dag> we do want to add hints to html output so we can search for them through Sphinx as well
15:45:55 <cybette__> resuming in a minute
15:46:35 * geerlingguy begins 60 second countdown... :D
15:46:56 <dag> felixfontein: I am probably confusing my work on ansible-doc for seealso, with the arg_spec PR
15:47:38 <bcoca> well, they can be related, if arg_spec starts being the source of docs
15:48:02 <cybette> this is a long minute... :P
15:48:16 <bcoca> mars minute its 1.3 earth hours
15:48:47 <felixfontein> luckily we don't live on mars
15:49:03 <bcoca> speak for yourself!
15:49:23 <bcoca> http://dilbert.com/strip/2008-05-08  <= in case anyone wonders about my 'mars math'
15:52:02 <felixfontein> it's hard to understand audio, there are always cut-offs
15:52:11 <bcoca> i've been thinking of adding to M() since now we document other plugins
15:52:40 * geerlingguy installs 2.7... proceeds to see a jillion deprecation warnings for all the packaging tasks that use with_item loops
15:53:04 <bcoca> was not aware of any with_ deprecation
15:53:16 <abadger1999> did that happen?  Thought we deferred that....
15:53:17 <bcoca> is this for squashing?
15:53:38 <geerlingguy> https://bin.acquia.com/?a861e2998d8add53#ULOqWXxlj271M92AbFYfjgKo50ovAp2iRywqMR1fmwk=
15:54:17 <bcoca> that does seem to be about squashing, didnt know we had deprecated that
15:54:18 <geerlingguy> basically if you use apt, pip, yum, package, etc. and with_items to pass in a list of packages it throws that warning now. It was a pattern I used for years until I discovered `name` takes a list
15:54:35 <bcoca> i thought we were going to tie ti to the deprecation/removal of with_
15:54:35 <felixfontein> the 2.7 rcs didn't complain about with_ loops
15:54:42 <geerlingguy> but there are sooooo many roles that still squash
15:55:00 <felixfontein> ah yes, there were warnings about squash
15:55:19 <dag> Do we already have a style guide for consistent Etherpad updates :)
15:55:22 <geerlingguy> example task more: https://bin.acquia.com/?e4c19daa1d4b26e6#YO33KxfQkAnsOpcBDYa1pKr1vbvqtBavjmXKjMnSQEQ=
15:55:30 <bcoca> found deprecation message
15:55:44 <felixfontein> what kind of webdev skills are you asking for?
15:55:53 <bcoca> last time i did 'webdev' 'div' was a 'new thing'
15:55:57 <dag> felixfontein: Javascript and CSS work
15:56:08 * geerlingguy guesses there will be a lot of new issues on ansible/ansible saying "I don't think I squash things, I just use with_items why is it showing this message" :D
15:56:12 <bcoca> <tr><td> is still in my muscle memory
15:56:33 <abadger1999> Better breadcrumbs
15:56:34 <dag> Every documentation includes the complete toc
15:56:34 <abadger1999> Better section headers on the module index pages
15:56:36 <abadger1999> Duplicate search facilities
15:56:42 <dag> which makes pages about 10x larger than they need to be
15:56:50 <abadger1999> (and of course, this integrates with sphinx somehow...)
15:57:01 <bcoca> breadcrumbs have never worked right
15:57:14 <dag> yeah, it is likely you need to upstream some of the things to Sphinx (especially the reference and search)
15:57:17 <bcoca> and toc should be link to 'grouping list' in plugins/modules
15:57:37 <felixfontein> the bluejeans image isn't really updating well
15:58:00 <bcoca> https://github.com/ansible/ansible/pull/35978
15:58:05 <bcoca> ^ deprecating squash
15:58:12 <bcoca> totally forgot this was there
15:58:52 <geerlingguy> chouseknecht needed to do like a 3 hour presentation :D
16:00:56 <bcoca> the search box was the original search, we added bottom search, but had to bring back box due to popular demand
16:01:02 <bcoca> ^ what i tried to say on bj
16:01:06 <bearrito> do the 2 searches behave differently? if not, is it much of a concern to have 2 places to search?
16:01:19 <bcoca> iirc, they are the same search
16:02:01 <abadger1999> they are different search
16:02:03 <bcoca> accidentally? we had a lot of people asking for it and dharma indicated he would
16:02:23 <abadger1999> a lot of people couldn't find the search.
16:02:38 <bcoca> ok, so current box is not what i discussed with him, we looked into it and could use same search with box
16:02:49 <bcoca> we even played around with some demo code
16:03:10 <bcoca> the site we use for the bottom search allows for input box
16:03:19 <abadger1999> If you have that demo code, that would be good.  We could get that integrated
16:03:41 <bcoca> i'll look for it, but it is probably in one of his branches
16:04:00 <bcoca> it was mostly a copy/paste from the sample code from the search provider
16:04:47 <bcoca> i had just assumed when it was added that it was a result of that .. didnt realize it was a 'sphinx accident' ... funny though
16:05:19 <bcoca> we had PR from user with drop down box for multiversion
16:05:26 <bcoca> but it was mostly hardcoded js
16:06:05 <bcoca> https://github.com/ansible/ansible/issues/45651
16:06:18 <bcoca> also looking for that code
16:07:29 <felixfontein> I ran make webdocs (though restricted to modules)
16:07:29 <dag> I run it very often these days
16:07:33 <bcoca> I do, when my heating is off
16:07:37 <felixfontein> haha
16:07:40 <dag> bcoca: it's much better now :)
16:07:42 <samccann> :-)
16:07:54 <felixfontein> yep, recent changes by dag make it much less heaty
16:07:59 <bcoca> yes, specially since we stopped building toc PER module page
16:08:03 <felixfontein> (at least if you're running it again)
16:08:09 <bcoca> still, its very cpu intensive
16:08:17 <felixfontein> yep
16:09:00 <bcoca> most of the time is spent in indexing and toc building, the templates seem to run pretty fast
16:09:10 <geerlingguy> qq - how will we manage multiple versions of a role for different projects?
16:09:21 <geerlingguy> Can I still install namespaced packages in a project subdir?
16:09:22 <felixfontein> please no iframes!
16:09:33 <bcoca> geerlingguy: i think you want #ansible-galaxy
16:09:43 <fxfitz> Yup ^^^
16:09:45 <jtanner> geerlingguy: no quick answer, but it's something we do need to figure out
16:10:03 <bcoca> <docs meeting here>
16:11:20 <bcoca> we use mostly jinaj2 templates to generate .rst for sphiinx
16:11:51 <bcoca> that step is normally fast, the rst => html (what sphinx does) is the slow/cpu hog part
16:12:39 <bcoca> i think im only hearing 1/2 the conversation
16:14:02 <felixfontein> rst => html in general should be pretty fast - if you just generate the page content. the problematic part is probably the TOC, handling links, etc.?
16:14:04 <cybette__> bcoca: there's not always someone speaking, but now Alicia is. can you hear her well?
16:14:37 <bcoca> no, and now i lost sound midsentence
16:15:06 <felixfontein> there seem to be long pauses from time to time
16:15:13 <cybette__> the pauses are there
16:15:37 <cybette__> but dropping off mid sentence is a problem. although we are sticking to the mics as much as possible here
16:15:39 <bcoca> mid word?
16:15:53 <bcoca> nvmd, bj dropped me
16:16:02 <cybette__> at least abadger1999 is taking great notes (thank you!!)
16:16:07 <felixfontein> how often are the devel docs regenerated?
16:17:28 <felixfontein> thanks!
16:17:40 <cybette__> felixfontein: once a day at around 2 or 3 in the morning central US time
16:23:37 <felixfontein> linting the examples sounds like a very good idea! (also spellchecking etc.)
16:24:04 <felixfontein> spellchecking is usually annoying
16:24:45 <felixfontein> if there's a exclusion list for spellchecking, there should also be one per module / module category
16:27:05 <bcoca> examples were not required to be valid yaml, it was freeform initially, i dont know if all were converted after we changed the restriction
16:27:40 <bcoca> some have plays, most others are just tasks, so linting might need to be very permissive
16:28:02 <felixfontein> hmm, that can actually be quite a problem
16:29:38 <bcoca> abadger1999: cannot find the search box code .. but it should be easy to recreate with example from our search provider
16:29:46 <abadger1999> k
16:30:06 <bcoca> also cannot find the contribution with a 'drop down' for versioning .. but htat was almost 3yrs ago
16:30:12 <abadger1999> I believe all examples were changed to valid yaml.  I know I converted some things to comments in examples to get around that
16:30:15 <felixfontein> I can see the terminal window
16:30:15 <bcoca> 'doc site version navigation'
16:30:47 <bcoca> abadger1999: i had to do same for updates, but dont know if we ever did for all existing
16:36:48 <digigrate_> Your breaking up up again.
16:37:43 <felixfontein> audio is not really working
16:37:53 <digigrate_> voice is breaking up^^^
16:38:01 <felixfontein> yep
16:38:14 <felixfontein> also with current speaker
16:40:06 <cybette__> how about now?
16:40:18 <felixfontein> cybette__: seems better
16:50:26 <geerlingguy> Is there going to be a kubernetes discussion today?
16:50:37 <geerlingguy> I thought I remembered seeing it in etherpad monday but it looks like not
16:50:52 <bcoca> https://etherpad.openstack.org/p/ansible-summit-october-2018
16:51:00 <bcoca> dont see it htere
16:51:11 <geerlingguy> k
16:52:25 <gundalow> Food is ready
16:52:56 <felixfontein> nice for you :) I have to cook myself
16:53:03 <gundalow> geerlingguy: no k8s session here. Though maybe we'll setup a working group
16:53:44 <geerlingguy> cool. I have a vested interest at this point as we're building two projects with Ansible + Kubernetes (instead of integrating Helm), aaaand I'm starting work on "Ansible for Kubernetes" :)
16:54:32 <geerlingguy> Kubernetes : scheduling infra awesomeness :: Ansible : automating infra awesomeness
16:54:52 <geerlingguy> It's funny, just like with Docker, there are many people who are like "why don't you just, like, log into servers and run commands and stuff?"
16:55:13 * geerlingguy Because we are enlightened beings. We don't log into servers, pfft.
16:55:18 <felixfontein> will there be (or is already?) an IRC channel for ansible documentation?
16:56:33 <felixfontein> sounds good!
16:58:01 <dag> #ansible-docs already claimed by @dharmabumstead, so we're good !
16:58:29 <felixfontein> thanks
16:58:39 <abadger1999> #endmeeting
16:58:53 <cybette> #endmeeting