14:02:05 <bexelbie> #startmeeting Fedora Docs
14:02:05 <zodbot> Meeting started Mon Jul 10 14:02:05 2017 UTC.  The chair is bexelbie. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:02:05 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
14:02:05 <zodbot> The meeting name has been set to 'fedora_docs'
14:02:15 <terminal__> .hello terminal__
14:02:15 <zodbot> terminal__: Sorry, but you don't exist
14:02:26 <terminal__> .hello t3rm1n4l
14:02:26 <bexelbie> .hello bex
14:02:27 <zodbot> terminal__: t3rm1n4l 'None' <stealthcipher@gmail.com>
14:02:30 <zodbot> bexelbie: bex 'Brian (bex) Exelbierd' <bex@pobox.com>
14:03:05 <rkratky> .hello rkratky
14:03:06 <zodbot> rkratky: rkratky 'None' <rkratky@redhat.com>
14:03:18 <bexelbie> #topic Roll Call
14:03:36 <terminal__> hey all
14:03:38 <pbokoc> I'm here but I'm in another meeting at the moment
14:03:47 <rkratky> hi
14:03:53 <bexelbie> Why have only one meeting when you can have two :)
14:03:54 <bexelbie> hi all
14:04:08 <bexelbie> # topic Agenda
14:04:14 <bexelbie> #topic Agenda
14:04:23 <bexelbie> I believe we have only three topics to discuss, but would like ideas for more if needed
14:04:27 <bexelbie> 1. Release Notes Status
14:04:31 <bexelbie> 2. AsciiDoc Conversion
14:04:42 <bexelbie> 3. bex's harebrained idea (asciidoc publishing asap)
14:05:01 <bexelbie> make that 4
14:05:06 <bexelbie> 1. Release Notes Status
14:05:12 <bexelbie> 2. Publishing for F26
14:05:17 <bexelbie> 3. Asciidoc Conversion
14:05:21 <bexelbie> 4. bex crazy pants idea
14:05:22 <bexelbie> any others?
14:05:45 <terminal__> nothing else that i know of
14:06:28 <bexelbie> ok, rolling with this
14:06:33 <bexelbie> #topic Release Notes Status
14:06:42 <bexelbie> rkratky, yruseva pbokoc ?
14:06:52 <rkratky> I got nothing
14:07:09 <bexelbie> are the items you assigned yourself going well? on target?
14:07:10 <pbokoc> I'm working through the ones I have assigned to myself, I'll be done in the evening
14:07:18 <rkratky> I mean, I§m all done w- mine. Reviewed and pushed,
14:07:54 <bexelbie> sclark ^^
14:08:17 <bexelbie> looks like we have about 22 release notes issues still open
14:08:31 <bexelbie> 1 unaassigned
14:08:33 <pbokoc> same for yruseva, she's working on pushing hers now
14:08:35 <bexelbie> #link https://pagure.io/release-notes/issues
14:08:57 <pbokoc> well, the issues being open are just there because people didn't close them yet. The commit log looks like we have a decent amount of content
14:09:07 <rkratky> I can take the unassigned one.
14:09:08 <pbokoc> I haven't gone through them and checked though, it's just my impression
14:09:14 <bexelbie> ok, so we may be good then
14:09:23 <bexelbie> I also haven't had time to review them all either :(
14:09:33 <bexelbie> hopefully we can move to close most of them
14:09:34 <pbokoc> more like "not totally screwed, maybe", but yeah :)
14:09:41 <bexelbie> pbokoc, exactly :)
14:09:53 <bexelbie> I don't know that sclark and randomuser are with us
14:10:00 <bexelbie> so we don't have their status, barring their tickets being updated
14:10:02 <rkratky> Oh, sgallagh commented that there's nothing to write about on the unassigned one. So I'm closing it, ok?
14:10:10 <rkratky> https://pagure.io/release-notes/issue/66
14:10:16 <bexelbie> rkratky, cool
14:10:20 <sgallagh> ?
14:10:32 <sgallagh> Oh right. Yeah
14:10:51 <bexelbie> so that's one more down :)
14:10:52 <pbokoc> hah
14:11:11 <bexelbie> ok, so we need to get the tickets reviewed for final closes and see if we need to prompt anyone
14:11:22 <bexelbie> I will try to do it, but I get on a plane in the morning so i can't promise
14:11:24 <pbokoc> sclark probably isn't, I think we're way out of his timezone, IIRC he was from Australia
14:11:32 <bexelbie> ok
14:11:35 * yruseva rushes through the door
14:11:46 <bexelbie> hopefully those will close too
14:11:48 <pbokoc> I'll try to go through everything too this evening if I have the time
14:11:49 <terminal__> it's 10 pm here in west australia
14:11:57 <terminal__> midnight over east
14:12:09 <bexelbie> I think we can move on to publishing - if there are no other comments.  I don't know that reviewing the tickets one by one here will accomplish much
14:12:43 <bexelbie> any objections ?
14:13:16 <bexelbie> #topic publishing for F26 release
14:13:32 <bexelbie> So, frankly, do we know how to do this?  I don't think I have rights, and even if I do, I don't know how to do it
14:13:40 <bexelbie> Can anyone other than randomuser do this?
14:13:43 <pbokoc> I've got nothing here - randomuser kinda sorta said he'll do it
14:14:14 <bexelbie> ok, conveniently I'll be in his timezone starting Wednedsday so I can lean on him a bit :)
14:14:16 <pbokoc> yeah, pkovar could in theory. The problem here is, you need a version of publican that last shipped with F18 - so you need a VM with that and publican
14:14:28 <bexelbie> pbokoc, ok
14:14:38 <bexelbie> I believe we should work under the belief that randomuser will do this
14:14:41 <pkovar> i don't have a working setup at the moment
14:14:53 <pkovar> if randomuser does, thats great
14:15:05 <pbokoc> at the same time, you need to push to the web.git repo on pagure, which you can't from that version of Fedora because of outdated ssl or something like that, pagure won't let you push
14:15:14 <bexelbie> Is this ok: #action randomuser to publish F26 Installation Guide, SysAdmin guide and Release notes
14:15:14 <pkovar> but we dont have a backup plan. do we
14:15:29 <pbokoc> pkovar, that's a nope
14:15:48 <bexelbie> with a follow up #action bexelbie to follow up with randomuser as he wasn't online today
14:15:54 <bexelbie> agreed? objections?
14:16:02 <pbokoc> I mean, randomuser doesn't have a working setup, the plan is, run publican inside the VM, and then push from outside it - mount the disk image and use the host to do the actual push
14:16:08 <pbokoc> and hope it works
14:16:21 <pbokoc> bexelbie, agreed, although it would be best if we could catch him today
14:16:25 <pkovar> ah, okay
14:16:26 <bexelbie> pbokoc, this is scaring me a bit -
14:16:33 <bexelbie> and #4 may change this some
14:16:36 <pbokoc> the release's tomorrow, so...
14:16:41 <pkovar> the due date is tomorrow?
14:16:42 <bexelbie> #action randomuser to publish F26 Installation Guide, SysAdmin guide and Release notes
14:16:50 <bexelbie> #action bexelbie to follow up with randomuser as he wasn't online today
14:16:54 <bexelbie> ok, I'll try to find him today
14:16:58 * bexelbie goes to SMS him
14:17:21 <pbokoc> bexelbie, eh, it's just a new ridiculous hack on top of another ridiculous hack that we've been doing for multiple releases now
14:17:29 <bexelbie> pbokoc, ok
14:17:32 <terminal__> :o
14:17:34 <bexelbie> I've SMS'ed him
14:17:49 <bexelbie> do we have more to cover in publishing? otherwise we can move on to getting out of this hack :)
14:17:57 <pkovar> yes
14:18:03 <bexelbie> pkovar, go ahead
14:18:18 <pkovar> also, i don't think anybody tested that the new location of the web repo actually works
14:18:28 <pkovar> i mean the scripting behind it
14:18:34 <pbokoc> oh yeah, that's a good point
14:18:35 <pkovar> as the old fedorahosted repo is now gone
14:18:45 <pkovar> so, even if we manage to push to pagure
14:18:55 <pkovar> the publishing might turn out to be broken
14:18:58 <pkovar> on the scripting side
14:19:05 <bexelbie> Is this aobut whether docs.fp.o actually updates?
14:19:06 <pkovar> just that you know
14:19:16 <pkovar> yes, pretty much
14:19:37 <pkovar> i couldn't test it due to pagure not talking to f18 anymore
14:19:59 <terminal__> ssh?
14:20:02 <bexelbie> pkovar, ok - my hope is randomuser can find a way through this but we need to actually be doing it at this point to patch the hack it seems
14:20:05 <bexelbie> monkeypatch if you will
14:20:19 <pkovar> yes, ssh, terminal__
14:20:32 <pkovar> alright, i'll be around too, tmrw but no promises
14:20:58 <bexelbie> I'll be on a plane most of the day for everyone - so I will not know what happened :)
14:21:12 <bexelbie> actually that isn't completely true
14:21:15 <bexelbie> I have a 7 hour layover
14:21:18 * terminal__ is freaking out
14:21:20 <bexelbie> so maybe I can scream fire with everyone else :)
14:21:44 <terminal__> pretty much
14:21:50 <bexelbie> ok, so any more fun with publishing or can we talk about getting out of this situation?
14:21:56 <yruseva> btw I *just* got a reply from one of the SMEs
14:22:11 <bexelbie> yruseva, ugh and hooray
14:22:29 <yruseva> bexelbie, he replied promptly, I asked late :)
14:22:41 <bexelbie> :D
14:22:51 <bexelbie> hooray and ugh then :P
14:23:04 <bexelbie> moving on to the next topic then, barring an objection
14:23:08 <bexelbie> #topic asciidoc conversion
14:23:22 <bexelbie> terminal__, has been working on converting the installation guide and system admininstration guide to asciidoc
14:23:27 <bexelbie> can you update us terminal__ ?
14:23:55 <terminal__> I haven't done a lot recently, although I tacked on that RHEL guide links stuff to the sysadmin guide as an issue
14:24:19 <terminal__> but i did run another attempt at converting the install guide
14:24:25 <bexelbie> how'd that go?
14:24:35 <terminal__> didn't make any difference
14:24:38 <terminal__> git says no change
14:25:17 <bexelbie> ok, does that mean we have it and think it is roughly good in asciidoc or tha thte conversion failed?
14:25:31 <pbokoc> I have to relocate, back in 2 minutes
14:25:46 <terminal__> it's not looking too bad, but there's some missing references
14:25:57 <bexelbie> ok
14:26:06 <terminal__> 2 figures, one table
14:26:32 <terminal__> not difficult, just got to work out where it's ref'd and insert the references
14:26:41 <bexelbie> ok
14:27:14 <terminal__> but there's another RHEL guide linked in the install guide
14:27:24 <terminal__> the kernel crash dump guide...
14:27:30 <linuxmodder> .hello
14:27:30 <zodbot> linuxmodder: (hello <an alias, 1 argument>) -- Alias for "hellomynameis $1".
14:27:46 <bexelbie> so of our big three (install, sysadmin, and release notes) it sounds like 2 are roughly converted to asciidoc, barring content updates (which I know pbokoc needed to make)
14:28:00 <bexelbie> terminal__, it sounds like we could leave those for now and not be any worse off than we were
14:28:07 <terminal__> pretty much
14:28:10 <bexelbie> if we wnated to prioritize the makrup change
14:28:10 <bexelbie> ok
14:28:21 <bexelbie> how long does it take to get the guides into "passable" asciidoc shape?
14:28:29 <bexelbie> like if we started on release notes tomorrow
14:28:34 <linuxmodder> bexelbie,  we did that last release cycle I'd recommend we not repeat and get further back
14:28:46 <bexelbie> linuxmodder, did what?
14:29:27 <linuxmodder> bexelbie,  this: [10 14:28:00] <bexelbie> terminal__, it sounds like we could leave those for now and not be any worse off than we were
14:29:36 <linuxmodder> but whatever not my call
14:29:51 <bexelbie> linuxmodder, those are the RHEL guide references - they have been there for a long time, I am not sure I follow what you are suggesting
14:29:57 <terminal__> depends on if there's anything else i missed in scanning the doc, but it's not as massive as the sysadmin guide...if there's no formatting issues, i'd just have to update the refs, and make sure the links work
14:30:13 <linuxmodder> again nvm Doesn't apparently matter
14:30:38 <bexelbie> linuxmodder, ok - I just didn't understand what you were suggesting - I am very happy to hear it and would like it to be considered if you wish to rephrase it
14:30:51 <linuxmodder> sysadmin unless someone worked on it recently had tons of outdated stuff in it
14:31:07 <bexelbie> This is true
14:31:09 <terminal__> I have converted it to asciidocs, and updated all the links
14:31:16 <bexelbie> Install Guide also needs some help
14:31:18 <linuxmodder> recently == < 2 months ago
14:31:31 <terminal__> it's still the same content though
14:31:41 <terminal__> does talk about F19 in the httpd section also
14:31:44 <bexelbie> linuxmodder, I agree that the content needs help
14:31:53 <terminal__> but then so does the apache web page
14:32:01 <bexelbie> but we haven't got content offers from people, so we are trying to get the format to one that will make contribution easier than DocBook
14:32:08 <bexelbie> your patches to content would be greatly welcome
14:32:18 <linuxmodder> the oldest references any of those should have is 24 tbh
14:32:34 <linuxmodder> any older refs are beyond outdated and irrelavant
14:32:40 <terminal__> yup...but it's not going to be ready by the time F26 drops
14:32:57 <terminal__> needs more people updating content and getting it updated
14:33:36 <terminal__> the guides themselves are all outdated, but at least in asciidocs it's a lot easier for new contributors
14:33:55 <terminal__> and potentially existing
14:34:04 <terminal__> the conversion process takes minutes
14:34:23 <bexelbie> so from a markup perspective, we could use what the new converted materials and probably have only a few display errors - is that a good read?
14:34:47 <bexelbie> from a content perspective we need help from people to get that shaped up and hope that asciidoc will help makethat easier for folks like linuxmodder to contribute
14:34:48 <terminal__> it's not terrible
14:34:51 <bexelbie> is that a good read/
14:34:52 <bexelbie> ?
14:35:26 <terminal__> markup looks fine for the most part...like i said a few missing references, the rest is content, and some formatting, but it's readable
14:35:35 <bexelbie> terminal__, cool
14:35:43 <bexelbie> any other markup conversion questions?
14:36:30 <terminal__> not from me
14:36:58 <terminal__> conversion is straightforward, it's just getting the doc updated and into shape after that that takes a little time
14:37:18 <bexelbie> ok then I'd like to see if I can get meeting-feedback on something
14:37:22 <bexelbie> #topic new publishing
14:37:31 <terminal__> not that i've got my head around the asciibinder stuff yet
14:37:33 <bexelbie> I suggest we try to move off of DocBook with F26 (even now)
14:37:34 <terminal__> but yeah
14:37:50 <bexelbie> It will not publish tomorrow, so tomorrow's publishing would be the last one with Docbook
14:38:09 <bexelbie> my proposal is to use asciibinder + ryanlerch's theme to create a new docs site
14:38:23 <bexelbie> we would use terminal__'s converted Sysadmin, install guide, and (not even started) release notes
14:38:31 <bexelbie> to publish those three on a new docs.fp.o that only has F26
14:38:44 <bexelbie> F25 and older would be on legacy.docs.fp.o (or something similar)
14:38:47 * randomuser shuffles in the back door
14:38:50 <bexelbie> this will cause the following problems
14:38:52 <terminal__> o/
14:38:56 <bexelbie> 1. There will be some display errors
14:38:58 <bexelbie> o/ randomuser
14:39:05 <bexelbie> 2. There will be no translations at first
14:39:18 <bexelbie> but we will get asciidoc publishing and conversion
14:39:26 <bexelbie> we will have the building blocks for automated publishing and testing
14:39:41 <bexelbie> on the translation front we aren't republishing regularly there now, so at least we will be consistent in languages
14:39:55 <bexelbie> we have a process that should be 85% clean now that we can use once translation verifies the new strings
14:40:03 <bexelbie> but it will have problems and we will have to fix them
14:40:15 <bexelbie> I really think we can do this as I have piloted most of it with budget.fp.o
14:40:18 <bexelbie> including publishing and staging
14:40:30 <bexelbie> any comments? Is this completely insane?
14:40:59 <randomuser> all sounds reasonable to me
14:41:07 <terminal__> sounds fine to me also
14:41:11 <pbokoc> we might not have another choice if web.git doesn't work
14:41:40 <terminal__> we need to get away from trying to rely on F18 and pagure to talk to each other :P
14:41:41 <bexelbie> terminal__, it is a huge request, but how long would it take to convert the Release notes once the last content was pushed into docbook?
14:42:13 <terminal__> conversion itself takes minutes...
14:42:21 <randomuser> wait, what? Why would we not be able to use pagure to publish docbook?
14:42:23 <bexelbie> but the cleanup you do is part of that
14:42:24 <terminal__> it's the verifying references and formatting after
14:42:35 <terminal__> yeah
14:42:47 <terminal__> how big are the release notes?
14:42:52 <rkratky> terminal__, I could help with the conversion.
14:43:01 <terminal__> I mean the sysadmin guide took me like 2-3 weeks
14:43:18 <terminal__> rkratky, that'd probably be necessary
14:43:19 <rkratky> RNs are relatively 'easy' content -- there shouldn't be much to fix.
14:43:34 <terminal__> esp. if the doc is large
14:43:49 <pbokoc> randomuser, well, have you tried that? With the workaround of pushing from outside the VM - we're concerned that the migration to pagure was just someone changing a .git/config to point to pagure and pushing, and nobody actually set it up so that the pagure repo replaces the fedorahosted one on the website
14:43:56 <pbokoc> we don't know for sure though
14:44:10 <bexelbie> do we think it is reasonable to say that we would have all content converted to asciidoc and ready to be used by 24 July?
14:44:18 <bexelbie> assuming no major problems
14:44:29 <bexelbie> with the goal of pushing translation strings that week and publishing by 1 August?
14:44:30 <pbokoc> also by the way, we have a better version of the script available internally, it has a lot of bug fixes. It's still not perfect, it still requires manual fixes to get the thing to even build, but it's likely to be better than the upstream version
14:44:36 <randomuser> pbokoc, i suspect that is the case, can check into it
14:44:56 <bexelbie> #chair randomuser pbokoc terminal__ rkratky
14:44:56 <zodbot> Current chairs: bexelbie pbokoc randomuser rkratky terminal__
14:45:01 <rkratky> bexelbie, I'm pretty confident Io could have the RN doc ready within one day of it being done. I tried with other docs.
14:45:01 <bexelbie> sorry forgot to do that
14:45:04 <bexelbie> anyone else I missed?
14:45:08 <pbokoc> I'll ask someone to get it somewhere publicly available, like on github or something
14:46:25 <terminal__> a large majority of the issues i see from docbook to asciidoc conversion seem to be the manual linebreaks everywhere
14:46:40 <terminal__> with the current docscii method that is
14:46:52 <randomuser> #action randomuser to update infra to pull from https://pagure.io/fedora-docs-web for publishing
14:47:03 <randomuser> that's the right repo, correct?
14:47:26 <pbokoc> randomuser, should be, yeah
14:47:26 <rkratky> terminal__, RNs have no (or very little) content other than regular text. The linebreaks shouldn't be an issue.
14:47:38 <terminal__> cool
14:48:28 <bexelbie> randomuser, is the current docs publish, literally them copying the content of fedora-docs-web to the servers?
14:49:34 <randomuser> bexelbie, last I checked the proxies would clone/pull the git repo
14:49:49 <randomuser> once an hour
14:49:55 <bexelbie> randomuser, cool that is how the new process works for budget.fp.o so we have even less work to do for a conversion :)
14:50:16 <randomuser> right, static content is fairly easy to distribute
14:50:19 <bexelbie> the budget site has two branches, prod and stg so we can separate stage from prod and make it easy
14:50:39 <bexelbie> my next goal is to get infra to do that on demand, and not via cron, but that is not a blocker
14:51:21 <randomuser> i'm curious but it sounds like a topic for another meeting
14:52:18 <bexelbie> randomuser, yes and one after we have converted
14:52:29 <bexelbie> I seriously believe we should commit that this is our last docbook publish
14:53:23 <randomuser> as you the person most responsible for providing the successor at this point, bexelbie, that's very encouraging to hear
14:53:51 <bexelbie> to do it, we need to commit to get content changes to land in the asciidoc that terminal__ is working on
14:54:09 <bexelbie> if we have support, I can work on building it out for success over the next two weeks
14:54:17 <randomuser> so what guides are ready to be published today in the docbook version
14:54:18 <bexelbie> I genuinely can't get it done sooner because of hte LATAM org fad
14:54:19 <randomuser> ?
14:55:50 <pbokoc> randomuser, nothing right now; I need to wrap up release notes and then I'll do a quick update for the install guide
14:56:26 <pbokoc> for the sysadmin guide I guess we'll just increment the release number and pretend it's up to date, I don't think there's much more we can do at this point in time
14:56:36 <pbokoc> I'll ping you when I'm done. Or e-mail the list or something
14:57:18 <pbokoc> (of course if anyone else wants to do the sysadmin guide I'll be happy to hand that over...)
14:57:41 <linuxmodder> i can do once overs but no cycles til next week
14:57:42 <randomuser> okay, I'll be around after work
14:57:55 <linuxmodder> wrt sysadmin & installguide
14:58:03 <pbokoc> and now I need to go find myself a meeting room for yet another meeting, yay
14:58:07 <randomuser> content reviews are helpful at any time, linuxmodder
14:58:41 <pbokoc> yes, but in that case I'd save your energy for after the asciidoc conversion, because that's going to need a whole load of once overs
14:58:50 <pbokoc> I'm out, l8r
15:00:36 <terminal__> I don't have a lot of spare time this week, works been pretty busy last few weeks, I may have a little time mid week atm but that's possibly about it
15:01:01 <adamw> hi, folks, I believe we have the room next? (qa)
15:01:09 <bexelbie> #endmeeting