18:00:18 <randomuser> #startmeeting
18:00:18 <zodbot> Meeting started Thu Aug 14 18:00:18 2014 UTC.  The chair is randomuser. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:00:18 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
18:00:20 <pbokoc> well played
18:00:27 <randomuser> #meetingname Fedora Docs Office Hours
18:00:27 <zodbot> The meeting name has been set to 'fedora_docs_office_hours'
18:00:51 <randomuser> seriously, I pulled some boxes out of storage that are apparently packed full of stank
18:02:47 <pbokoc> in a few years you'll love it. You'll be driving home after a day of hard work and you'll smell it from half a mile away - and you'll be all "Mmm. Smells like home."
18:04:21 <randomuser> oh, nice, we have new badges
18:04:44 <randomuser> https://badges.fedoraproject.org/badge/commis-cookbook-i
18:06:07 <randomuser> there is a series - docs-writers can manually assign these after helping someone with a recipe
18:06:42 <randomuser> we should probably keep track of how many badges have been assigned to a person, somehow
18:07:32 <pbokoc> doesn't that appear in the person's profile?
18:07:47 <randomuser> pbokoc, no, it doesn't
18:08:25 <randomuser> like, if someone has contributed four articles, we'd have to know that so the next one would get the 5 article badge
18:08:33 <pbokoc> right
18:08:49 <randomuser> but there *is* a pseudo-metadata block at the top of the template
18:09:22 <randomuser> so in theory we could `grep -R fas_id en-US/|uniq -c`
18:09:38 <randomuser> as seen in https://git.fedorahosted.org/cgit/docs/fedora-cookbook.git/tree/en-US/Creating_GPG_Keys.xml
18:10:15 * jsmith is back :-)
18:10:42 <jsmith> randomuser: Mind making me a chair?
18:10:49 <randomuser> certainly
18:10:53 <randomuser> #chair jsmith
18:10:53 <zodbot> Current chairs: jsmith randomuser
18:10:58 <randomuser> #chair pbokoc
18:10:58 <zodbot> Current chairs: jsmith pbokoc randomuser
18:12:34 <randomuser> hmm.. make that `git grep`
18:14:19 <pbokoc> well, as long as we make sure to maintain the comment on each recipe
18:14:40 <randomuser> yes, i think it's important that we do that
18:14:55 <randomuser> we can also use that to generate a contributors list later on
18:15:28 <pbokoc> well, are we expecting people not familiar with Git to contribute to the Cookbook?
18:16:14 <randomuser> pbokoc, yes, that's why the badge is set up to be manually awarded
18:16:51 <randomuser> we want people to submit articles in various formats and stages of completion and get help with the rest of the process
18:16:58 <pbokoc> yeah, I know, but it's not just that. If we required everyone to just send patches, that would take care of the contributor list, but it still wouldn't work with the badges
18:17:23 <pbokoc> because git shortlog -sn tells you someone has x commits, but it doesn't distinguish if it's 5 commits to a single recipe, or 5 separate recipes
18:17:34 <randomuser> right
18:17:46 <randomuser> we have a different badge for commits, anyway :)
18:17:58 <pbokoc> yeah, I'm just nitpicking
18:18:41 <randomuser> I did ask about author vs committer in the git log, for patches, but fedmsg doesn't do that right now
18:18:55 <randomuser> and it sounded non-trivial to implement
18:19:11 <pbokoc> also on an unrelated note, in which beat would you put RPM update? https://fedoraproject.org/wiki/Changes/RPM-4.12
18:19:26 <randomuser> uhmm
18:20:15 <pbokoc> yeah, my words exactly
18:20:22 <randomuser> that's fairly low level, maybe wedge it in with the systemd stuff?
18:20:39 <pbokoc> I think I'll just lump it in the "Development" pile
18:20:44 <randomuser> or the kernel stuff, or something similarly distant from the user?
18:21:17 <randomuser> pbokoc, iirc it has soft requires or suggests or similar - if you mention that make sure you also mention our packages don't do that
18:21:49 <pbokoc> systemd... there's the Boot beat, that doesn't really fit, and then there's the System Daemons beat, and that's not much better
18:21:53 <pbokoc> roger
18:23:48 <pbokoc> in earlier RNs there was some content related to RPM in Server Configuration Tools...
18:24:05 * randomuser notes that Croberts has a fresh badge coming if he ever finishes his cookbook submission
18:24:22 <pbokoc> like "dnf greets Fedora" or "yum-presto merged into yum"
18:24:32 <randomuser> pbokoc, i think we went through the same 'meh, i'll just throw it in $here' process in the past
18:24:37 <randomuser> do as thou wilt
18:24:47 <pbokoc> ok then
19:31:08 <jhradilek> Uff. Finally home.
19:37:26 <jsmith> jhradilek: It's about time...
19:37:30 * jsmith ducks and hides
19:38:30 <jhradilek> :))
19:39:17 <jhradilek> randomuser: Is now a good time to talk about crazy ideas on how to make it easier for new contributors to access our documentation?
19:49:24 <randomuser> jhradilek, absolutely
19:50:03 * randomuser grabs jsmith' sandwich while he is 'missing'
19:50:30 <jsmith> randomuser: That's fine... I didn't know you like peanut butter and cucumber sandwiches :-p
19:51:24 <randomuser> jsmith, i haven't tried that, actually - but peanut butter and pickle, peanut butter and banana, peanut butter and bacon, or any combination of those - yes
19:51:28 <jhradilek> randomuser: I already mentioned it to jsmith in Prague and yesterday, I had a chance to talk about it with Matthew Miller. I mentioned that one of the problems we have with our cookbook and other documentation is that in order for it to be available, we need to publish it, which is... slow and very difficult at the moment.
19:51:58 <jsmith> Yup...
19:51:59 <randomuser> yeah - when the new system gets off the ground that part should be faster
19:52:15 <randomuser> ...not wiki fast, but better for *us* doing it now
19:52:26 <randomuser> anyway, go on
19:52:43 * jsmith has to run pick up his son
19:53:00 <jhradilek> randomuser: I am not sure if you are familiar with Jenkins, but basically, it could be an answer to this. We do have it configured in our office and we even have a pretty front end that shows you a list of books along with a link to their repos, links to HTML and PDF builds, a build log, and a button to rebuild the book immediately.
19:53:10 <pbokoc> peanut butter cucumber sandwich... http://i.imgur.com/1SxptUZ.gif
19:54:17 <randomuser> jhradilek, interesting... and that triggers builds on commit?
19:54:17 <jhradilek> randomuser: This would give us two things: access to the latest development version of our documentation automatically built from the source code whenever somebody pushes a commit, and a nice little overview of our currently developed books with links to their repos.
19:54:51 <jhradilek> randomuser: Yeah, Jenkins watches repositories and whenever it detects new changes, it creates a new build.
19:55:12 <randomuser> i *think* other fedora things are using it
19:55:27 <jhradilek> randomuser: I have an engineer on my team who could help set it up for documentation and fine tune the user interface for our needs.
19:55:42 <jhradilek> randomuser: Yeah, Matthew mentioned that Fedora does have a Jenkins instance and that we should use it.
19:55:51 <pbokoc> yeah, it either builds automatically, or you can schedule a build manually at any time if you don't want to wait for the autodetection to kick in
19:55:59 <jhradilek> randomuser: My question is if this is something you would like as our leader. :))
19:56:05 <randomuser> ha!
19:56:10 <randomuser> hell yeah, how can I help
19:56:15 <pbokoc> https://fedoraproject.org/wiki/Jenkins@infra
19:57:26 <jhradilek> randomuser: We need to compose a list of books we want to have there. I will ask Pavel, the engineer I mentioned, to provide instructions on how to set up Jenkins for them.
19:57:44 <jhradilek> randomuser: And then we need a server where we could run the front end.
19:57:56 * randomuser nods
19:58:15 <randomuser> the build part seems like it would be very straightforward, the output side less so
19:58:19 <jhradilek> It's not resource hungry and have minimal dependencies.
19:59:04 <randomuser> getting a vm instance shouldn't be a problem
19:59:18 <randomuser> we might even have one for docs we aren't using
19:59:51 <jhradilek> randomuser: If you find one, would it be possible to give me a temporary access to it, so that I could start setting things up?
20:02:13 <randomuser> jhradilek, I don't want to commit to more than I can take on right now, and I'm kinda already there - maybe we should take this to #fedora-infra and point the you at someone else?
20:02:25 <randomuser> or if we can catch nb, he can help
20:02:43 <randomuser> err.. #fedora-admin is what I meant
20:02:47 <jhradilek> randomuser: I understand! I'll talk to nb. :)
20:03:24 <randomuser> in a week or two, I'll be completely moved and have a lot more time. from 1+hr commute to 5min commute
20:04:07 <randomuser> jhradilek, so what happens if the build fails?
20:04:30 <randomuser> should we start doing silly things like a pre-push xmllint hook, too?
20:05:17 <jhradilek> It turns red, provides a link to the build log, and offers you the last successful build.
20:05:29 <randomuser> cool
20:06:20 <jhradilek> Well, once we have our books in Jenkins, we could start using test automation. I am working on an open-source project for that. :)
20:06:26 <randomuser> i regret it every time, but sometimes I break master :P
20:07:04 <jhradilek> Heh. You know what helps? :)
20:07:21 <jhradilek> Whenever somebody breaks my book, they owe me a beer. :)
20:07:36 <randomuser> nice
20:10:24 <randomuser> jhradilek, I'll go with that. Whenever I break the build on my own book, I'll treat myself to a beer before continuing
20:10:34 <randomuser> this is sure to improve the quality of my work
20:10:48 <pbokoc> that's what I do
20:11:06 <jhradilek> pbokoc: You know I shouldn't know that, right? :)
20:11:07 <pbokoc> I can't say it makes me break things less though
20:11:10 <pbokoc> but it's delicious
20:11:19 <pbokoc> jhradilek, you always knew :-D
22:08:09 <randomuser> well, this has been fun
22:08:13 <randomuser> #endmeeting