23:02:04 #startmeeting Docs Project Meeting - Agenda: http://fedoraproject.org/wiki/Docs_Project_meetings 23:02:04 Meeting started Wed May 26 23:02:04 2010 UTC. The chair is sparks. Information about MeetBot at http://wiki.debian.org/MeetBot. 23:02:04 Useful Commands: #action #agreed #halp #info #idea #link #topic. 23:02:11 #meetingname Fedora Docs 23:02:11 The meeting name has been set to 'fedora_docs' 23:02:18 #topic Roll Call 23:02:18 * sparks 23:02:23 * bcotton is here 23:02:23 * rudi_ is here 23:02:25 * jjmcd 23:02:26 * ke4qqq is here 23:02:33 * laubersm is here 23:02:46 * stickster 23:03:44 * sparks thinks nb will be along soon. 23:04:08 Okay, lets get right to it! 23:04:13 #topic Follow up on last week's action items 23:04:19 oops 23:04:20 * nb here 23:04:24 Looks like I had both action items from last week. 23:04:36 nb I'm assuming you have all the access you need, now. 23:05:05 yeah, we decided not to remove people manually, but to file a bug for infra to do it i think 23:05:12 nb: Okay. 23:05:20 and the second one was also completed. 23:05:37 Anything else from last week? 23:05:49 i have a few, but i'll mention them later in my area of the meeting 23:05:52 Ok 23:05:56 Moving on, then 23:06:01 #topic F13 Release, What went right and what went wrong. 23:06:21 Okay, just wanted to review the release and try to capture "lessons learned". 23:06:35 First, everyone did an excellent job! 23:07:02 So in final review, how did we do on updating Guides for F13 content? 23:07:07 We did a lot of changing in process and the guides got out the door in time ad on schedule. 23:07:12 Lots of help especially from lots of new folks on the RN conversion, as well as many stepping up to take guides was a huge win 23:07:28 jjmcd: Agreed. 23:07:45 stickster: I think that's something that we haven't tried to measure. 23:08:10 The "how relevent are the guides to the release" question. 23:08:26 I guess we will find out based on the responses we receive. 23:08:39 sparks: I think I'm looking for some sort of affirmation from the owners 23:08:46 heh 23:08:47 Like ke4qqq for the IG, laubersm for the UG 23:09:05 I have a pretty good idea of where we stand with RN, since I worked a bit with jjmcd et al. 23:09:20 * jjmcd was disappointed he never found what he needed in the virtguide, but then he's weird anyway 23:09:26 The SG added more content. Since most of it isn't release/version specific it would be hard to answer that question. 23:09:40 IG and IQSG include substantial rewrites for F13, hence getting them to L10N so late 23:09:41 sparks: But you would say it's pretty accurate for F13 then 23:09:45 I think the UG went well . 23:09:46 stickster: If the RNs aren't covering F13 content then we have problems. 23:09:46 stickster: honestly I felt like we stepped up with a lot of guides, and a lot of guides I felt like we still HEAVILY relied on rudi - and he did great work, but I fear raptors will hunt him down 23:09:47 :) 23:09:51 I had lots of help from new folks (and oldie bu goodies) and I think the UG went well. 23:09:53 sparks: Exactly! 23:10:13 Yeah, the SG is accurate for F13, IMO. 23:10:37 * Gearoid is here 23:10:46 * rudi_ hopes that once publishing is fully automated, the workload will be spread much more evenly across all docs and L10N members :) 23:10:47 ke4qqq: Yeah, I agree. rudi_ is a great asset but we shouldn't rely solely on his work. 23:10:50 rudi_: We obviously want more writers, but where did you feel strained as far as guide production? 23:10:53 I needed rudi a good bit but most of those things are in the wiki now (now if I could find them with search) 23:11:13 ianweller: ^^^ 23:11:53 The Accessibility Guide was completely revamped from previous versions and was much more relevent but still incomplete. 23:12:09 I found the procedures much easier than I feared ... having not done much before but getting to use the newer tools of git and publican (without having to merge po files separately) 23:12:13 I'm hoping to get Gearoid to take over that guide and help do some of the research. 23:12:13 stickster -- just the mechanics of building and publishing; but that was partially a factor of changing methods and it still felt like a lot less work with Publican+git as opposed to hand-coded PHP+CVS :) 23:12:21 The last minute churn on docs.fp.o was a delta, but going forward is going to be good 23:13:03 So it seems to me that (1) content-wise, we did pretty well; (2) production-wise, things were easier than F12, and even better in future now that everyone is learning the new tools 23:13:04 * jjmcd got the zero day release notes to the site late, because I didn't want to add further to the confusion 23:13:04 jjmcd -- yeah, it was unfortunate, but remember last time, we took flak for changing the site layout in the middle of a release :) 23:13:11 yep 23:13:30 Substantial changes are always going to expose issues that nobody thought of previously 23:13:31 The churn on docs.fp.o caused quite a few problems although I think it's going to be a good thing going forward. 23:13:34 There's no good time to make big changes. That being said, right before release is probably less good than others 23:13:35 And BTW, got the published pretty much from the wiki 23:13:38 yeah 23:13:43 or just hadn't joined the dots on 23:13:50 sparks, yeah I hope to be of more help from now on, finished work next week and off all summer! 23:14:19 i thinik the docs site launch went ok, no downtime 23:14:22 * stickster would motion that future big site changes should happen in accordance with infrastructure freeze 23:14:23 * ianweller is here 23:14:32 stickster, technically we did, we got a freeze exception 23:14:36 sparks: what am i looking at 23:14:38 kudos to the admin folks for remedying the issues that we put before them. 23:14:40 but yeah, i'd suggest doing them before the freeze next time 23:14:41 nb: Yeah, we did 23:14:51 Well, this should be the last change of this nature 23:15:05 ianweller: laubersm made a comment about search in the wiki. I'd like to work with you/others to try to come up with a better solution. 23:15:07 For a while, yes! but never say never... :-) 23:15:14 Even when we get publishing fully automated, the site structure itself doesn't change 23:15:15 ianweller, I was griping about wiki search 23:15:19 sparks: so would i. 23:15:24 I tell you what, having nb around #f-d a lot was a real key to making it all happen 23:15:25 Oh yeah; I was talking about the foreseeable future :) 23:15:32 rudi_: ;-) 23:15:33 sparks: it's one of my things for this summer. we'll see where it goes 23:15:35 +1 jjmcd 23:15:41 ianweller: Is there a Google thing or something else we can use? 23:15:42 nb is worth his weight in gold :) 23:15:50 rudi_, exactly 23:15:52 * stickster a bit distracted, have to finish wrapping mrs. stickster's b-day presents as we go here 23:16:21 stickster, you're supposed to have the jewelry store do that 23:16:47 stickster: need a finger to help with the bow? 23:17:07 stickster, duct tape holds anything together 23:17:36 Has anyone received any feedback (that hasn't made it to the list) about the docs.fp.o frontend? 23:17:49 Can you wrap duct tape *in* duct tape? 23:17:58 why not? 23:17:58 stickster, yes 23:18:02 Some comments on the channel about broken links 23:18:06 been there done that 23:18:07 we got lots of them 23:18:12 jjmcd: There are bunch of those. 23:18:31 we set up redirects for most of the high-profile ones 23:18:35 Has anyone heard specifically about the experience of people in non-en-US locales? 23:18:39 The one that came up on the docs and trans lists was particularly interesting. 23:18:39 the cookies thing on the list which is fixed already 23:18:39 and if you know of others, i can do so 23:18:44 Are they finding their translated content where it appears? 23:19:05 stickster: They are finding their content but if they don't have content they aren't finding anything. 23:19:23 So I'll be forced to ask the question... 23:19:23 I noticed that too when I first looked at the site... 23:19:49 I noticed a kind of weirdness today that I'll talk with rudi about later. Very minor, tho 23:19:53 Why is it bad to publish partially translated guides when the alternative is a no-translated guide? 23:20:01 I wanted to see what lang the UG was in and couldn't find a way to do it other than go through each language (or the source of course) 23:20:09 sparks, some teams really hate that 23:20:13 sparks -- because it's ugly and unprofessional 23:20:30 all or nothing is a better presenation 23:20:42 Yeah, some people think it encourages people to join in, but people who are actually reading the docs aren't always bought into that notion yet. 23:20:45 laubersm -- we'll have a report for that available soon 23:20:55 but if I go under es and I don't see a particular guide I don't even know that it availabe in en... 23:21:01 then the non-English reader will have nothing which I don't agree with. 23:21:08 I thinkg that is what they were commenting on as well 23:21:11 that sould be corrected 23:21:14 should 23:21:30 stickster -- I think the resolution we're hashing out on the mailing list will maybe give us the best of both worlds there 23:21:30 laubersm: They were 23:22:00 yes, seems like the ml is heading to some sort of concensus on that 23:22:01 It's a simple fix... but not the best when it comes to the %age of completed translation. 23:22:08 Hopefully seeing the list of books in the index, some translated some not, will provide some stimulus to get involved 23:22:18 Without actually publishing any piebald content :) 23:22:36 I will have an admonition in place within hours 23:22:40 perhap nontranslated links in red 23:23:01 And will work today with Jeff to see what we can do about titles 23:23:19 jjmcd, most of the langs have the guide name traslated... just being in english is hilite enough... 23:23:21 This is really excellent feedback for Publican :) 23:23:43 I noticed that Jeff fixed our lang.css bug really fast. 23:23:47 laubersm, good catch, altho it doesn't leap out in all langs 23:24:10 * stickster likes rudi_'s use of 'piebald' 23:24:15 stickster -- yeah, using your solution :) 23:25:24 rudi_: Just read your message to L10N. Looks good to me. 23:25:26 :) 23:25:46 Okay... anything else on this topic before moving on? 23:25:46 Thanks 23:26:10 Let's move along, then. 23:26:15 #topic * 23:26:28 shoot 23:26:31 stickster: how do you undo? 23:26:36 #undo 23:26:50 ^^ 23:26:51 cool 23:27:05 #topic Schedule for F13 for F14 23:27:07 nb: Thanks 23:27:13 you didnt undo yet sparks 23:27:19 it didnt work because i'm not chair 23:27:36 Oh well... lets leave it be, now. 23:27:37 :) 23:27:42 ok :) 23:27:50 Okay so we did pretty well with the schedule this time around. 23:27:59 Anyone know of anything we should put on the schedule for next time? 23:28:15 most of the schedule actually made sense this time. Still a few touch-ups 23:28:36 jjmcd: Can you make a list of the touch-ups and provide them to poelcat? 23:28:51 I figured we would uncover them in the F14 meetings 23:29:04 Most are negotiated wth L10N anyway 23:29:12 #info If there are touch-ups needed, include them as actionable tasks with specific start and end dates for poelcat 23:29:23 heh 23:29:23 Yeah; I have to sit down and think through a few things 23:29:49 rudi_, exactly. Hard to recall where we adjusted when the sked didn't make sense 23:30:03 jjmcd: rudi_: The list would show us... if people had those conversations on the list :-) 23:30:05 Okay. We can come back to this at a later week. 23:30:07 But at least this time we weren't pushing the rpm before we built it 23:30:17 heh 23:30:28 I probably need ryanlerch's help to draw a picture :) 23:30:37 dot is your friend 23:31:56 #action rudi_ and jjmcd to come up with updates to the schedule for F14 release by 30 June 2010. 23:32:02 or there abouts 23:32:14 or a start... anyway. 23:32:19 Okay, anything else? 23:32:45 * poelcat will have a draft f14 schedule out soon 23:32:58 jjmcd: please send your feedback based on it 23:33:06 poelcat, wilco 23:33:09 i'm using the same logic and spacing as f13 23:33:21 we're using logic? 23:33:26 It was WAY closer than F12 23:33:36 which was better than F11 23:33:48 so we're going in the direction we want to go 23:33:51 poelcat: Thus, once again, the iterative schedule proves its value :-) 23:33:53 Can we automate things based on the F15 schedule? 23:34:22 * sparks dreams of hands-free documentation. 23:34:27 * poelcat remembers calling stickster on the phone several years ago to construct the first detailed docs schedule 23:35:14 :-) 23:35:20 sparks: I think we can probably move on 23:35:23 poelcat: Thanks for what you do... it really helps. 23:35:26 Okay... moving on! 23:35:28 +1000 23:35:33 #topic Release Notes 23:35:43 Heh 23:35:45 jjmcd: So when are you going to have the F14 RNs completed? 23:35:55 whenever poelcat says 23:35:59 :D 23:36:13 Basically, RN's were pretty much nominal 23:36:39 I didn't get all of stickster's fixes into d-p-r before the z-day, so there was a little manual touchup on that one 23:36:39 I think I said this before but... it felt like there were fewer 0-day changes. Do you agree? 23:36:49 Very few 23:37:04 What can you attribute that to? 23:37:06 I think part of that was the slip at the end allowed us to catch up a bit 23:37:29 Im slowly getting the release thing figured out, still a few screw ups there but next time will be good 23:37:39 And really, it isn't a big burden 23:37:40 * sparks was hoping it was because of increased communication with those out in the trenches doing the work. 23:37:58 so I've been encouraging L10N to bump me when they want a lang added 23:38:09 sparks, I think that played a part 23:38:11 and i think having jjmcd in packager now helps probably 23:38:17 I'm not sure, but it felt to me like we got fewer contributions this time from developers. We should see if ianweller can whip up something to tell us about the history of all the beat pages, and see who made changes along the way. 23:38:18 On a related note, I've worked out a way to further streamline localisation of the TN for next time 23:38:25 Well, now I have to work instead of sparks 23:38:49 stickster, some of that, but the release notes in the feature pages are getting better too 23:39:02 jjmcd: That's good to hear! 23:39:02 rudi_, cool 23:39:19 stickster: hmm. 23:39:20 * ianweller opens the cage to the python exhibit 23:39:20 I didn't play as a big a role in pulling content in this time, so my view may be askew. 23:39:25 python! go! 23:39:26 jjmcd: Are people using the RN entry in BZ? 23:39:36 Both 23:39:46 in documentation and fedora 23:39:51 no 23:39:58 which is a little confusing, I've been moving them 23:40:16 I mean... there is a text field in every BZ ticket where people can flag a ticket for the RNs. 23:40:28 Oh, that one 23:40:35 sparks: That's really for development changes that deserve a release note 23:40:35 I think maybe only one of those 23:40:41 As opposed to a bug in the docs themselves 23:40:54 Oh wait 23:40:56 Yes. I was wondering if people were using that text field or nto. 23:40:58 not 23:40:59 I totally misunderstood you 23:41:04 Yeah, I bet we had fewer than a dozen all told 23:41:17 I thought you were saying "man, why are they doing things weirdly on BZ when there's this NEAT FLAG?!?" 23:41:22 heh 23:41:32 once again, IRC ruins good human understanding :-) 23:41:46 I need to rant on about that some more. The bugs are REALLY helpful 23:41:47 * sparks doesn't speak human well anyway. 23:42:59 Those who provide bugs are generally more descriptive than the notes in the features. A lot of the features were good 23:43:00 * stickster must go afk 23:43:10 but some it is really hard to figure out what's the point 23:43:27 stickster: Have a good evening 23:43:40 because the developer is so wrapped up in it, it's obvious to him 23:43:46 ya 23:43:54 * sparks thinks he's been following that ticket 23:43:58 take care stickster 23:44:07 be good to the birthday girl 23:45:26 sparks, I was referring to the feature pages 23:45:31 ok 23:45:32 oh 23:45:41 * stickster back 23:45:42 that particular bug had dozens of good things if it's the one I'm thinking of 23:45:50 And still a few to follow up on 23:46:28 * sparks notes we are running short on minutes 23:46:37 jjmcd: Ya.. 23:46:43 Anything else for the RNs? 23:46:54 Nothing else really major, lets move on 23:47:11 #topic Guide Status 23:47:28 So everyone published their guides, right? 23:47:43 * laubersm published the UG 23:48:04 * sparks needs to update some of his guides. 23:48:45 UG had 4 other langs at 100% (nl, it, es, and uk) 23:48:57 sparks -- talk to me before that -- there are some special problems :) 23:49:06 rudi_: ya 23:49:25 looks like ru is still being worked on and is now at 89% I will keep an eye on it and add it when it gets done 23:49:31 Excellent! 23:49:45 yeah, some L10N teams are still going strong 23:49:47 On that note... 23:49:56 does everyone have what access they need with the new groups? 23:50:09 we really need to find "someone" to take a look at Transifex performance issues 23:50:23 L10N was really stymied when working on the bigger guides 23:50:26 diegobz is one of the admins i think? 23:50:40 I'm not sure where the bottleneck is 23:50:55 Maybe the server that it's running on doesn't have enough grunt for the job 23:51:05 rudi_: Define stymied. 23:51:15 rudi_, more iron is always good 23:51:18 They literally can't upload translations 23:51:26 really?!? 23:51:30 transifex runs on the appservers iirc, and i think they have pretty good specs, /me pings mmcgrath 23:51:31 yeah 23:51:38 mmcgrath, ricky jds2001 ping 23:51:45 So I've had to work with them on committing directly through Git 23:52:02 Which is not optimal from a workflow point of view 23:52:05 ya 23:52:13 * nb is seeing if he can get someone that might know here 23:52:24 It exposes them to exactly the kind of collisions that Transifex protects them from 23:52:34 There's some discussion on the L10N thread about this 23:52:38 L10N list sorry 23:52:54 eh? 23:53:05 Hey smooge 23:53:06 * abadger1999 looks in 23:53:21 I can only be here for a short while. what can I help you with? 23:53:24 smooge, abadger1999 we're discussing performance issues with transifex and the larger guides 23:53:38 * nb is not sure who in infrastructure knows much about transifex 23:53:41 so this is the first I've heard about it, when did it start 23:53:42 I was just noting that L10N has reported major hassles with the Transifex interface with the larger guides 23:54:12 About a week ago 23:54:14 abadger1999 I think knows a lot more about it than me.. my knowledge is pretty (ok very) limited 23:54:27 When teams started working on the IG 23:54:50 Look for a thread in trans-list "Problems with Installation Guide" 23:54:53 Were they slowness issues or correctness issues? 23:55:12 However, there's been fairly constant background kvetching about performance 23:55:16 abadger1999, iirc they were "transifex doesnt get it committed" 23:55:20 Slowness 23:55:28 is what someone was saying 23:55:30 Slowness to the point where the connection times out 23:55:40 before they can submit their work 23:56:00 And lots of regular "page not found" stuff 23:56:09 But I can't quantify that 23:56:24 But that was going on for ages before things reached a head with the installation guide 23:58:09 k... we'll need to ask upstream transifex I think. 23:58:19 glezos: ^ If you're around. 23:58:31 I've just asked if any translators are around to join us 23:58:34 I have to head out. Sorry I wasn't much help 23:58:44 smooge, thanks anyway 23:58:44 glezos: If you're not we'll send you a note. 23:58:54 NP smooge -- thanks for stopping by and for all your help over the last few days 23:59:00 Much appreciated :) 23:59:14 abadger1999: rudi_: I recall glezos saying there were a significant number of fixes in 0.8.x Tx that address known problems 23:59:33 The move from 0.7.x -> 0.8.x is a lot less drastic than what we just did (0.5? -> 0.7) 23:59:49 stickster: I think they're getting ready for 1.0 as well... 23:59:49 Anyway, glezos can surely fill us in. 23:59:55 abadger1999: I wouldn't be surprised 00:00:01 * stickster eof 00:00:10 When is the best time to upgrade tx? 00:00:18 Should we be doing it around now if possible? 00:00:29 abadger1999: Maybe in a couple of weeks? 00:00:43 * sparks thinks some L10N members are still working. 00:00:54 abadger1999, several teams seem to be actively working on the F13 guides at the moment... 00:00:56 Yeah; still some activity. 00:01:05 Can we fix a date to check again? 00:01:32 Say in two weeks, we reappraise whether there's significant translation activity still happening? 00:01:36 Okay -- If you can give me a date when we can start workin gon it, that would be good. 00:01:50 And then we need to get someone to actually test the upgrade on stg and deploy :-/ 00:02:10 #action rudi_ to come back in two weeks with whether or not there is signficant translation activity. 00:02:14 rudi_: Sounds good... ping us in two weeks to look at getting started? 00:02:25 Awesome. 00:02:27 Thanks abadger1999 -- we'll update you either way 00:02:36 * noriko sorry coming late 00:03:09 Okay, anything else on this? 00:03:32 Sorry noriko -- we were just discussing some of the problems that translators have had with Transifex performance during docs translations 00:03:52 what problem? 00:04:18 Not being able to upload work; pages timing out; the stuff that's been discussed on trans-list 00:04:52 Did you experience any of that yourself? 00:06:08 rudi_: once yes, but it had been fixed, afaik 00:06:18 Ah OK 00:06:48 So the problems might only still be manifesting in the big books like the Install Guide and Virt Guide 00:07:47 rudi_: I can't tell much. it happens sometime and back to normal. no one tells us what was or is wrong. 00:08:11 Okay... let's move on 00:08:17 Thanks noriko! 00:08:50 #topic git repo conversion 00:08:54 nb: You around? 00:10:13 yikes 00:10:50 * laubersm didn't realize we wer having a double session tonight... 00:11:06 Okay... I was hoping that nb could talk about the changes that were made last week. 00:11:13 #topic git repo conversion 00:11:26 * nb here 00:11:32 Hi! 00:11:37 All the git repos now require docs-writers to commit 00:11:41 nb: talk about what you did... and what needs to be done 00:11:47 except for docs/web.git which requires docs-publishers 00:12:01 and I'm getting ready to send out emails about moving the other docs under docs/ 00:12:06 and a few renames along the way 00:12:13 uitstekend 00:12:26 I didn't do them at first because it will mess up people if they do not change their .git/config file when the move happens 00:12:40 although i'm going to see if symlinks work with git, then i could symlink the old location to the new location 00:12:54 all the git* repos for docs are now obsolete 00:13:13 Excellent! 00:13:14 we will eventually make them "go away", although you can self-remove yourself from them now if you want, or if you are admin, you can remove everyone from them if you want 00:13:20 as well as all the svn* groups 00:13:50 I didn't touch svnpublican because it isn't a docs group 00:13:56 publican is a redhat thing i think 00:14:19 We may eventually get it so that all members of docs-publishers can commit and build the actual packages 00:14:36 it is possible, but will require a few code changes, pkgdb-client doesn't support using groups besides provenpackager at the moment 00:14:51 any questions or aynthing? 00:15:14 Did everyone get all the access they need? 00:15:14 nb: And the pkgdb-server may need some updates too (at least the webui) 00:15:23 nb: pong 00:15:31 jds2001, unpong 00:15:43 jds2001, it was about the transifex slowness we discussed earlier in here 00:15:46 nb, so far, everything I've tried has worked 00:16:03 some people may not have gotten sponsored into docs-publishers 00:16:09 i didn't have a real good list of who to add 00:16:13 but we have several sponsors 00:16:14 nb -- everything working fine here so far -- many thanks 00:16:16 .sponsors docs 00:16:17 nb: Sponsors for docs: bcotton ianweller @jjmcd @ke4qqq @kwade @laubersm @nb @pfrields @quaid rlandmann @sparks sradvan tsagadai 00:16:18 .sponsors docs-writers 00:16:21 .sponsors docs-publishers 00:16:21 nb: Sponsors for docs-writers: bcotton @jjmcd @ke4qqq @laubersm @nb @pfrields @quaid @rlandmann @sparks @sradvan tsagadai 00:16:24 nb: Sponsors for docs-publishers: @jjmcd @ke4qqq @laubersm @nb pfrields quaid @rlandmann @sparks sradvan 00:16:28 contact those people if you need in a group you aren't already in 00:16:41 * laubersm had to sponsor stickster_afk for publishers (after stickster_afk had sponsored me for docs way back when) 00:16:45 That's a good number of sponsors! 00:16:48 laubersm, lol 00:16:58 laubersm to the rescue! 00:17:03 sparks, yeah, i think that was our idea :) 00:17:14 Yep! 00:17:54 * nb EOF (I think, unless someone else has someting for this topic) 00:18:04 The website changes went ok, except for all the broken link issues 00:18:07 Anyone have anything else for nb or htis topic? 00:18:10 which we added rewrites to 00:18:13 most of them 00:19:01 nb: Thanks for the info and all your hard work! 00:19:05 all i have is applause 00:19:53 Okay, thanks nb... 00:19:55 moving on! 00:20:13 #topic All other business 00:20:26 Anyone have anything they need to talk about before we close? 00:21:25 Okay... Well, thanks everyone for coming out! 00:21:37 I appreciate everyone's work during the F13 release! 00:21:38 thanks sparks good meeting 00:21:47 #endmeeting