14:00:31 #startmeeting docteam 14:00:32 Meeting started Wed Apr 23 14:00:31 2014 UTC and is due to finish in 60 minutes. The chair is annegentle. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:33 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:35 The meeting name has been set to 'docteam' 14:00:44 howdy 14:00:45 hello 14:00:46 Agenda is here: https://wiki.openstack.org/wiki/Meetings/DocTeamMeeting#Agenda_for_next_meeting 14:00:47 hi 14:01:06 a tom appear 14:01:08 hey fifieldt Sam-I-Am gpocentek AJaeger! 14:01:19 #topic Action items from last meeting 14:01:48 Summer used action for topic last meeting it looks like 14:02:04 I finished this one: annegentle write blog post linking to OpenStack Documentation Contributions survey 14:02:15 #link http://justwriteclick.com/2014/04/10/tearing-down-obstacles-to-openstack-documentation-contributions/ 14:02:41 how's the feedback on that? 14:02:41 it could use more tweets, I'll keep it open until the Summit (or a few days before to give me time to make nice graphs) 14:03:22 we're at 32 responses 14:04:03 is that about what you were expecting? 14:06:04 I was hoping for more... it's interesting. 14:06:21 1/3 don't have an XML setup, just under 1/3 don't have gerrit 14:06:21 any other ways to get more responses? 14:06:27 i'm sure you have it figured out :P 14:06:56 I'd really appreciate the doc team using G+ and Twitter to get the word out further, but yeah I've done the newsletter twice, blog post to planet, and included it in What's up Doc 14:07:15 i dont really have a following of openstack folks in g+ 14:07:20 and... not on twitter 14:07:31 The other action was: Ajaeger1 to send backport script to openstack-docs mailing list and it's done, thanks AJaeger 14:07:47 Sam-I-Am: yeah fair for the amount of time you've been in the community :) 14:08:01 maybe that'll get fixed during the summit! 14:08:15 not too many people know me outside of the docs team 14:08:21 if we had 150 contributors to docs, that's a good return on the survey, but it'd be nice to get non-contributors to respond as well 14:08:42 and I do think the responses are not just from us the docs team, so that's good, just want more data 14:08:52 #topic Icehouse release 14:09:13 I feel so good about the Icehouse release thanks to everyone's hard work. and that we didn't have to lock to stable yet 14:09:31 I wanted to take a few minutes to do a post-mortem. Let's start with What went well? 14:09:54 install guide :) 14:10:01 +1 14:10:12 +1 14:10:15 wait, -1!!! 14:10:16 Yes! 14:10:25 CLI Ref was amazingly smoothly executed 14:10:29 +1 14:10:34 also good 14:10:38 and the config ref took a ton of work but the results are great 14:10:43 +1 14:10:55 fifieldt: that's more +1s than i've ever seen you give :) 14:11:02 +0 14:11:08 What can we improve or not do next release? 14:11:17 I have to say the Ops Guide felt like a bit of a distraction 14:11:23 the config ref can still be improved 14:11:49 the install guide needs some structural/appearance help in most chapters 14:11:50 there's been lots of work on the ops guide before the release 14:12:31 how do people feel about the api side of things? 14:12:40 gpocentek: yes on the config ref, for you that may have been a lowlight since it was a tough road 14:12:41 in terms of process improvements, we need a way to ensure editorial reviews without distracting from technical updates 14:12:52 fifieldt, Diane does a great job there - but it's mainly her 14:12:52 dianefleming: +1 to that 14:13:02 dianefleming: +1 14:13:08 indeed, all hail diane 14:13:15 thanks !! 14:13:19 * fifieldt wishes there was some kind of tooling to assist 14:13:27 AJaeger: there was some invisible-to-gerrit editing for the User Guide really close to the release, too 14:13:33 because it's a big job 14:13:43 but i need to take some time to write up some of the "how-to write API docs" wiki pages 14:13:46 fifieldt: dianefleming had an idea of two workflows for technical edits and quality edits 14:13:47 invisible to gerrit editing? 14:13:48 so others know how to contribute 14:14:07 AJaeger: oh just that our technical editor marked up a PDF and Karin entered the edits. So not quite "invisible" but not inline in gerrit I mean 14:14:21 AJaeger: though I did edit in gerrit too 14:14:30 yes, my idea was, in gerrit, people vote on technical content, and then we have some sort of flag that indicates that an editorial review is also needed 14:14:38 annegentle, yeah. Figured that one out - Karin's edits are great in general, just was a lot of edits at once 14:14:39 dianefleming: yeah the API repos came up in the TC meeting yesterday 14:14:42 and then that change gets into another queue 14:14:44 AJaeger: yeah 14:15:01 and has an editorial review - but it wouldn't be necessary for all changes - 14:15:01 dianefleming: just asking what the plan is for those -api repos 14:15:14 oh! yes - the api repos 14:15:19 dianefleming: (TC members were asking) 14:15:27 my plan is to still move them back to the project repos - 14:15:38 once i complete that API guide 14:15:48 I sense that v3 Compute was a bit of a bug-number-raiser for this release 14:15:52 i will make that a priority in next few weeks now that icehouse is going out 14:15:55 dianefleming: yep that sounds right 14:16:16 database-api is moving to trove - see https://review.openstack.org/#/c/87807/ 14:16:18 I also think that continuing to document nova-network and neutron is not great for this release. 14:16:26 AJaeger: yep 14:16:34 what do you mean (neutron and nova-network)? 14:16:38 AJaeger: and we want that for all the projects, it just takes a while 14:16:38 in terms of API docs? 14:17:01 dianefleming: well if neutron was completely feature-finished to replace nova-network, we wouldn't have to keep documenting both 14:17:08 and here's a gating patch for trove https://review.openstack.org/#/c/88050/ to work with the move... 14:17:13 dianefleming: with each release that's the goal but it slips each time, that's all 14:17:28 ha, "all" -- it's kind of a big deal. But it means our release docs have to reflect reality. 14:17:28 annegentle, if you pass this around, give both reviews as example what needs to be done 14:17:34 AJaeger: good idea 14:17:38 i need to think about it a little more - there might be a quicker path than what I've planned - 14:17:55 Any other highlights or lowlights for Icehouse? 14:17:58 i was waiting to move those repos until I finished other work 14:18:09 amazing teamwork 14:18:27 amazing work by Andreas on build gates 14:18:28 the install guide was quite the eye-opening project 14:18:28 dianefleming: definitely 14:18:34 +1 14:18:44 plus one to AJaeger for alllllllll the build work 14:18:47 this is an awesome team 14:18:58 love working with y'all 14:19:00 +1 on the awesome team! 14:19:05 Sam-I-Am: +1! 14:19:17 i get home from work and look forward to my 'second job' 14:19:27 ;) 14:19:53 :) 14:20:53 Sam-I-Am: that describes me back in 2009 or so, seriously! 14:21:02 working on FLOSS Manuals 14:21:04 anyway 14:21:09 you were a dentist? 14:21:14 heh 14:21:26 http://en.flossmanuals.net/ 14:21:38 #topic O'Reilly Ops Guide release 14:22:26 With amazing publishing prowess, O'Reilly will be sending us an epub today or tomorrow for us to post 14:22:38 I need to write a blog entry for OpenStack.org 14:22:48 and Tom and I may get a chance to talk to some media about the project 14:23:00 yayyy 14:23:02 There will be copies at the Summit 14:23:06 Cool! 14:23:26 I still am in awe of their publishing power. I literally landed the back cover quote Friday afternoon and it's going to be printed this week. GEEZ 14:23:28 woohoo 14:24:06 Any questions on it? Launch plan seems solid and the Foundation is happy to launch this! 14:24:41 Any special guidelines for doing work and approving it going forward? 14:25:04 AJaeger: good question, I have a question in about when they are landing the index entries for me to backport 14:25:16 AJaeger: so that backport will make the source look much much different 14:25:22 AJaeger: due to all the index entries 14:25:33 we can "freeze" the tree a bit to get the index in... 14:25:35 AJaeger: but even once those are in, reviews are the same 14:25:48 AJaeger: yeah there's not a ton of activity on it and I've been watching like a hawk 14:26:00 * AJaeger tries to add you to reviews ;) 14:26:34 though the poor sheepdog contributor who wants his listed but the cinder folks just don't think it's tested enough... anyway 14:26:42 AJaeger: :) 14:26:45 I've been doing more reviews for what little that is worth... I can try and do more of those, myself. ;-) 14:26:56 that's been helpful sld 14:26:56 sld: awesome and super helpful 14:27:04 more reviewers = good 14:27:07 heh. 14:27:11 fo sho. 14:27:12 specially tech reviews 14:27:22 #topic Doc contribution survey - please post everywhere 14:27:28 heh I already talked about that 14:27:45 moving on 14:27:48 #topic Doc tools news 14:28:14 I noticed this week that the api-site wasn't building what I thought should be built on draft-docs.openstack so AJaeger is already addressing 14:28:21 thanks AJaeger -- anything else to add? 14:28:47 oh and dianefleming has requested a new release of the clouddocs-maven-plugin 14:28:55 that'll get us the PDF links for the API references 14:28:56 yes, just for api-site 14:28:58 annegentle, gpocentek is constantly improving the autohelp part ;) 14:29:10 nice! high5 gpocentek! 14:29:14 :) 14:29:24 otherwise tools work just fine... 14:29:28 so many awesome people 14:29:37 yes we won't update the version for openstack-manuals pom.xml files 14:29:46 just for api-site pom.xml 14:30:22 Any questions on doc tools? 14:30:29 * annegentle pauses to think for a moment 14:30:47 seems like translations are all tidy and sorted, right? 14:31:00 oooo question about that.... 14:31:07 oh and in that blog post I linked to above, please take a look at the requirements listing for our doc tools 14:31:09 is there any way to have those auto-approve themselves? 14:31:28 sld: oh good question. I think some projects have gate tests that allow that? AJaeger do you know? 14:31:48 I think a quick human check might be a good idea, to see if it's horribly broken or not 14:32:02 I agree with Tom 14:32:03 annegentle, regarding translations I improved the tooling as well: http://jaegerandi.blogspot.de/2014/04/changes-for-importing-translations-to.html 14:32:05 k 14:32:20 #link http://jaegerandi.blogspot.de/2014/04/changes-for-importing-translations-to.html 14:32:20 nice 14:32:22 sld you mean: translation going in without approval? I'm not aware of that 14:32:23 but those things have a TON of files they change....anything in particular to look for? 14:32:32 AJaeger: yeah, exactly. 14:32:37 the machines, sld, they're taking over :) 14:32:40 hehehe 14:32:42 resist! 14:32:44 naw 14:32:45 heh on the bot needing to sign the CLA! 14:32:46 what I look for 14:32:54 is files being removed 14:32:55 blanked 14:33:00 completely redone 14:33:03 new files added 14:33:12 sld, we had for quite some time translation jobs failing and now they finally work 14:33:16 I look at the summary of files 14:33:20 and then click a few random ones 14:33:25 just to see what's gone onm 14:33:26 lol AJaeger 14:33:32 normally it looks like small translation progress 14:33:37 which means good = approve 14:33:41 I look at one or two files and see that they are fine - and then check files with a strange number of changes... 14:33:42 fifieldt: normally i look to see down the whole list that they are changes, specifically...not addition or deletion. 14:33:50 k. 14:33:51 yeah 14:33:55 that kind of thing 14:33:58 k. 14:34:06 sld, so it's more a sanity check that they are fine. We now have some extra checks to see that they work. 14:34:13 though, one day, we should automate :) 14:34:23 when skynet goes live 14:34:33 that's a great set of pointers 14:34:42 #action everyone look at the requirements listing in http://justwriteclick.com/2014/04/10/tearing-down-obstacles-to-openstack-documentation-contributions/ and give feedback on the openstack-docs mailing list 14:34:43 the checksyntax gate tests the syntax of locale files 14:34:49 lol skynet 14:34:57 #action annegentle to send post about doc tools requirements to openstack-docs mailing list 14:35:17 ok anything else on docs? 14:35:21 er tools I mean? 14:35:52 #topic Summit doc track decisions 14:35:56 #link http://summit.openstack.org/cfp/topicstatus 14:36:17 that link shows how "oversubscribed" some of the projects are for track talks. We're fine at 5 proposed for 4 slots 14:36:32 #link http://summit.openstack.org/ 14:36:48 Take a look at the Doc topics, any we should combine or further open up? 14:37:59 they all look like good sessions 14:38:11 yeah, they do... 14:38:13 yeah 14:38:31 I wonder if we can open up the Ceilometer one to all recently integrated projects doc plannign? 14:38:45 Seems like they all should be asking good questions like those 14:39:21 +1 14:39:26 Should we combine Install Guide with HA active/active? good or bad? 14:40:07 I think more focus on HA stuff should be 'on hold', so to speak, until this is done - https://wiki.openstack.org/wiki/Neutron/L3_High_Availability_VRRP 14:40:17 annegentle: -1 on combining those 14:40:30 yeah, I think both better separate too 14:40:35 We have no real contributors to the HA guide ;( 14:40:40 sld: oh interesting 14:40:42 yup 14:40:47 yeah, sld has an interesting point there 14:41:09 I can help with HA stuff, but with VRRP on the horizon (pun intended. LOL), we should see what is up with that. lol 14:41:12 the only things i can sort of see combining are "patching the docs process" and "continuous publishing and automation" 14:41:15 I'm not sure about Percona leading the HA discussion in a docs track? Anyone else have insights? 14:41:41 if they want to write tonnes of things, we should encourage that :) 14:41:41 Sam-I-Am: yeah those could be combined 14:41:50 but, yes, it would be better to be more spread 14:42:16 fifieldt: hm my spidey sense/intuition does not tell me they want to write, they want writers :) 14:42:26 Percona? ...a database is a database is a database... lol 14:42:27 well, that makes it easier :) 14:42:40 fifieldt: to? 14:42:44 annegentle, maybe that's the session to cut then? 14:42:50 reject ;) 14:42:52 fifieldt: ah I see 14:42:57 * annegentle is slow :) 14:43:00 not at all! 14:43:07 * fifieldt is just not making sense 14:43:11 as usual :D 14:43:16 lol 14:43:18 * AJaeger does not understand fifieldt 14:43:22 did you try rebooting? 14:43:29 coffee? 14:43:35 I need two +2s to reboot ;) 14:43:40 but keep spelling it wrong 14:43:44 or using the wrong tags 14:43:53 spelling counts! 14:44:17 also wanted to be sure you all know about the two docs cross-project sessions 14:44:20 /me thankfully remembers how to spell atrocious 14:44:59 #link http://junodesignsummit.sched.org/event/626a1e21babaf30d98973f5eb7402fcf#.U1fR5uZdU0w 14:45:09 #link http://junodesignsummit.sched.org/event/19381e6ad48e05abc9099eb7ff956231#.U1fR8uZdU0w 14:45:23 I'll schedule the other four today (or by the end of the week for sure, gotta get that blog post done) 14:45:33 * annegentle time blocks 14:45:47 ok 14:45:48 you should try non-blocking time, its more efficient 14:45:50 #topic Installation guide 14:46:04 the good news... it works. 14:46:06 #link https://wiki.openstack.org/wiki/IceHouseDocTesting 14:46:10 so few bug reports 14:46:13 * fifieldt is scared 14:46:19 So all I wanted to talk about here is how we'll know when to cut stable/icehouse? 14:46:23 i'm in the process of expanding my lab and testing the release versions of everything 14:46:25 woohoo on working!! 14:46:38 however, there is still much work to be done 14:46:40 Wow Sam-I-Am 14:46:58 just to set a bound: pre-summit at the latest? 14:47:03 annegentle: my question is also your question 14:47:17 and we should base it on what's left to do 14:47:33 we need to prioritize 'need to do' vs. 'want to do' 14:47:35 #link https://launchpad.net/openstack-manuals/+milestone/icehouse 14:47:38 Sam-I-Am: or just say pre-summit 14:47:47 and backport from there 14:47:50 some things can wait for juno, others we can backport 14:47:56 it's kind of freeing in some ways :) 14:48:09 Sam-I-Am: I haven't checked lately, is OVS-non-ml2 removed? 14:48:14 its gone 14:48:24 * annegentle sings "Let It Go" from Frozen at the top of her lungs 14:48:27 buuuut... 14:48:55 i noticed that rhel 7 is following in fedora's footsteps by using "ens###" for network interface names rather than "eth#" 14:49:02 the diagrams and stuff reference "eth#" 14:49:12 ewww 14:49:27 if we're going to make the guide compat with rhel 7, i am considering "generifying" our description of network interfaces 14:49:31 or at least trying 14:49:47 this issue affects the arch, basic os config, and neutron chapters 14:50:38 arrr. 14:50:46 fifieldt: did you notice the "ens###" on your fedora box? 14:50:57 nope, it was using eths 14:51:01 whats more interesting is "ens###" is not consistent among boxen 14:51:10 it kind of chooses based on pci bus ID or something 14:51:49 i think we can either generify the guide or use notes to mention ens-style distros 14:52:27 this will be the first big change for juno 14:52:29 I believe 14:52:58 so you dont want to cover it in icehouse? i suspect it only affects people using rhel 7, which isn't released yet... hence why we haven't seen bug reports. 14:53:01 so I think the longer we wait to cut stable/icehouse, the longer Sam-I-Am's holiday is :) 14:53:11 heh 14:53:12 lol 14:53:22 what holiday?! 14:53:27 Sam-I-Am, could go either way for me :) 14:53:36 * fifieldt defers to other's judgement 14:53:37 does anyone know when rhel 7 will release? 14:53:50 rh likes to be secret about it 14:54:05 even if it's released soon, there's alway a delay before adoption 14:54:10 I've heard something about a RC these days, so shouldn't be long. 14:54:22 secret secret 14:54:24 betas are out 14:54:32 it should take too much work to generify the docs (he says cautiously) 14:54:37 I'd make the changes in juno 14:55:04 well, tell you what. i'll work on a patch. if it happens after we cut stable, we can always backport it. 14:55:19 might not be a pretty backport 14:55:32 Sam-I-Am: I'm fine with Icehouse being "not 7" 14:55:42 reflect the reality of 4/17/14 14:55:57 annegentle: thats one way to look at it 14:56:06 ZDNet reckons Q3 2014 14:56:07 Sam-I-Am: yeah we do timed releases 14:56:10 and of course, if "ens###" ends up in our cut, its not a big deal 14:56:14 Sam-I-Am: so it makes sense to me anyway 14:56:21 the other thing about rhel7 is mariadb 14:56:25 which the guide doesn't cover at all 14:56:30 so there is another road block of sorts 14:56:58 yeah that has to go in juno 14:56:59 i think i can agree with targeting rhel 7 for juno 14:57:04 ayup 14:57:27 next thing... diagrams 14:57:35 oh yes 14:57:44 i had a plan of adding diagrams to chapters which add optional components 14:58:05 Sam-I-Am: that's a great plan, want to do it in Juno tho? Or really want for Icehouse? 14:58:08 for example, the cinder chapter would have the arch diagram with a block storage node added 14:58:14 I'm all for incremental improvement to avoid burnout! :) 14:58:25 just to give people an idea of what's running on the new node 14:58:32 plus network connectivity 14:59:08 * fifieldt likes diagrams 14:59:13 annegentle: i'm hoping they wont take long since i have 'templates' of sorts already 14:59:22 just need to add the little service boxes inside of them 15:00:19 fifieldt: me too... visualization good. 15:00:32 ok sorry our time's up 15:00:34 this also follows along with me producing some sort of 'template' for diagrams 15:00:41 Better carry on in #openstack-doc 15:00:45 #endmeeting