19:00:01 #startmeeting FESCO (2010-03-30) 19:00:01 #meetingname fesco 19:00:01 #chair dgilmore notting nirik skvidal Kevin_Kofler ajax pjones cwickert mjg59 19:00:01 #topic init process 19:00:02 Meeting started Tue Mar 30 19:00:01 2010 UTC. The chair is nirik. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:00:03 Useful Commands: #action #agreed #halp #info #idea #link #topic. 19:00:07 The meeting name has been set to 'fesco' 19:00:08 Current chairs: Kevin_Kofler ajax cwickert dgilmore mjg59 nirik notting pjones skvidal 19:00:10 Afternoon 19:00:17 greetings, starfighter 19:00:20 * skvidal is here 19:00:25 Present. 19:00:30 pjones: wow, that's a blast 19:00:35 * nirik is ready to face Zur and the Kodan armada. 19:00:38 I need to see if they have that on dvd 19:00:42 they do. 19:00:56 * nirik notes we are missing cwickert and notting in channel. 19:01:04 notting is on vacation 19:01:18 so I think he's unlikely to be hopping in from his cruiseship 19:01:19 skvidal: there's even a 25th anniversary edition 19:01:53 ajax / dgilmore: around? 19:02:04 pjones: looking to see if it is captioned, too 19:02:46 ok. I guess we do have quorum, so we can go ahead and get started... 19:03:02 skvidal: Subtitles: English, French, Spanish 19:03:26 pjones: interesting - the one on netflix only says french 19:03:28 skvidal: collector's widescreen edition seems to be english and french only 19:03:34 ah ha 19:03:40 okay - so, back to the meeting :) 19:03:42 * dgilmore is here 19:03:52 ok, first some followups / status updates. 19:04:02 #topic #347 tor is not compliant with Fedora guidelines 19:04:08 .fesco 347 19:04:09 nirik: #347 (tor is not compliant with Fedora guidelines) - FESCo - Trac - https://fedorahosted.org/fesco/ticket/347 19:04:29 The tor maintainer has made correct changes for rawhide/devel... but doesn't want to push those to stable releases. 19:05:09 I can try and talk him into it further, I think he mis-understands what we are asking for... 19:05:14 Those changes should be pushed to F11/F12/F13 ASAP as they fix some reported bugs, in addition to making the package (hopefully) compliant to our guidelines. 19:06:08 I think he's thinking we want the entire init script re-writen with no lsb stuff, but really all we want is chkconfig instead of init_install or whatever crazy lsb thing it's using. 19:06:17 okay two questions 19:06:46 1. compelling him to do backport b/c of a policy issue seems like a bad choice - thinking about all the pkgs out there that aren't compliant 19:07:01 2. if this is a high-impact item can any provenpackager hit them? 19:07:08 #info bug https://bugzilla.redhat.com/show_bug.cgi?id=532373 19:07:20 nirik: I've been reading the bug 19:07:56 Yeah, I agree with both. I think it would be good to fix... but of course it needs testing to confirm it's not going to break things... 19:09:08 nirik: and I'm still a little fuzzy on how this is high-impact 19:09:39 I was suggesting to him that he push this as part of any other update that needs doing in the statble releases... 19:10:39 so, 1) wait a week and I will talk to them, 2) ask a provenpackager to just do it, 3) just close this and let stable releases wait? 19:10:49 I'm for 1 and 2 19:10:51 actually 19:10:53 so something like 19:11:10 1. wait a week and talk to them - and float the idea of 2. some provenpackager doing it 19:11:21 +1 to that. 19:11:23 skvidal: that wins 19:11:43 sounds fine to me. 19:11:51 would you like to update the bug? Or should I? 19:12:00 the bug or the ticket? 19:12:14 either I guess. 19:12:33 sure, why not? 19:12:42 (i'm here) 19:12:56 ajax: was that intended as a vote? ;) 19:13:01 not really. 19:13:41 but skvidal's idea sounds sane 19:13:51 * skvidal will update the ticket and bug 19:14:00 #agreed Will ask maintainer to please push the update to other releases, and look at getting a provenpackager to do the update if nothing happens in a week 19:14:12 #action skvidal will update ticket/bug for tor 19:14:20 #topic #351 Create a policy for updates 19:14:35 I have some info here, but think we might be best just defering a week for when notting is back. 19:14:49 +1 to defer for notting 19:14:53 +1 19:14:58 +1 19:14:59 Basically the changes in pkgdb/bodhi shouldn't be very hard... abadger1999 doesn't have much time to work on it so we may need to get someone to do patches there. 19:15:02 the world ain't gonna change in a week 19:15:25 defer +1 19:15:27 notting was working on figuring the critpath loading stuff. 19:15:37 #agreed Will defer this another week. 19:15:44 #topic #356 'distribution' vs 'general' in bugzilla 19:15:56 This is just pending hearing about sorting it to the top of the list. 19:16:06 no word back on that yet that I know of. 19:16:10 .fesco 356 19:16:11 nirik: #356 ('distribution' vs 'general' in bugzilla) - FESCo - Trac - https://fedorahosted.org/fesco/ticket/356 19:16:20 So defer & move on. :-) 19:16:28 yep 19:16:30 yep. Just informational. 19:16:39 notting was looking in to that, IIRC. 19:16:44 so, you know, defer for notting 19:16:49 https://bugzilla.redhat.com/show_bug.cgi?id=576320 19:17:07 #agreed defer to hear back from bug 19:17:17 #topic #359 Exception request for Virt Bridged Networking feature 19:17:20 .fesco 359 19:17:25 nirik: #359 (Exception request for Virt Bridged Networking feature) - FESCo - Trac - https://fedorahosted.org/fesco/ticket/359 19:17:40 https://fedoraproject.org/wiki/Features/Shared_Network_Interface 19:17:58 basically a done feature that was not added to the feature process on time. 19:18:27 I don't see any good that can come of /not/ granting this. 19:18:35 likewise 19:18:39 Yeah 19:18:47 yeah, it's done, it's in, it should be touted. 19:18:56 * skvidal doesn't like the word 'tout' 19:19:01 but other than that - seems fine 19:19:15 sad that NM integration isn't there tho 19:19:39 there's always next time 19:20:06 +1 to accepting this feature. 19:20:10 ok, I think thats +6... 19:20:21 #agreed Late Feature is accepted. 19:20:31 #topic #361 Units policy 19:20:36 .fesco 361 19:20:37 nirik: #361 (Units policy) - FESCo - Trac - https://fedorahosted.org/fesco/ticket/361 19:20:48 basically ubuntu has implemented some kind of units policy... 19:20:48 I believe we should use units 19:20:59 we have been asked if we want to do something similar or don't care. 19:21:17 KDE has been following that for a while. 19:21:20 I think MiB and whatnot are the stupidest thing that has happened since the advent of MS-DOS, and that's really saying something. 19:21:28 (that's right, I think it's dumber than java.) 19:21:31 * skvidal wonders what it means for fesco to adopt this? 19:21:34 I think we should be consistent. however i think that most of the time people really don understand the differences between base2 and base10 numbers for the units 19:21:43 as i said on the ticket: i don't think it's something we need to address. 19:21:46 pjones: +1 on MiB being silly as goose nuts 19:21:54 MiB is correct. 19:22:02 MB is a million bytes. 19:22:06 upstream (at least one) is asking us... 19:22:09 We need to stop the abuse of MB to mean MiB. 19:22:19 Kevin_Kofler: nice retcon. 19:22:20 * skvidal falls into the doesn't give a crap category 19:22:34 anyway, +1 for agreeing not to take this subject up. 19:22:37 though I'm pretty sure i'll never say MiB b/c it makes me think of Will Smith 19:22:43 pjones: It has always been an abuse of "mega-" to use it for 1024². 19:22:55 and that makes me think of the song 'Miami' and I don't want that 19:22:59 It's just that there wasn't a better term, there is now. 19:23:02 nirik: "at least one"? 19:23:07 Kevin_Kofler: if you can live with that fiction, I guess I'm happy for you. 19:23:36 ajax: well, the person who filed the ticket said one of the packages they maintain asked them, so they were asking us. 19:24:10 not to be impolite, but rahul tends to make mountains of molehills. 19:24:16 k means kilo which in bytes means 1024. M means Mega which is 1024 of those. As the episcopals say, "as it was in the beginning, is now, and ever shall be." 19:24:17 I think we should be consistent with what KDE does and ban using kB/KB, MB, GB, TB for 1024^n and recommend the use of KiB, MiB, GiB, TiB everywhere. 19:24:40 I think we should be consistent and ban KDE everywhere ;) 19:24:47 Kilo means 1000, not 1024. 19:25:01 children. we're not being asked to define terms here. 19:25:02 I personaly dont get the point of MiB etc 19:25:05 * skvidal thinks we should tilt at windmills 19:25:06 in seriousness, though, I still vote for not taking this point up. 19:25:12 And one reason why KDE uses the KiB etc. terms is that they're unambiguous. 19:25:15 we're being asked whether fedora should have a policy about it. 19:25:15 its not how its taught in schools 19:25:19 When you read "KB", you don't know which it is. 19:25:24 Kevin_Kofler: no, they're just differently ambiguous. 19:25:24 The Ubuntu units policy does not represent the usage anywhere else in the universe 19:25:31 ok, so how many 'We should just not have an opinion' votes is that? 19:25:36 When you read "KiB", you know for sure what is meant. 19:25:42 Kevin_Kofler: no I don't. 19:25:56 I dont either 19:25:58 KiB is always 1024 bytes. 19:25:59 So I think us adopting it is unhelpful, and I don't think there's anything else to standardise on 19:26:01 nirik: i count three ( me pjones skvidal ) 19:26:07 MiB 1024², GiB 1024³ etc. 19:26:11 If individual projects want to have their own style guides then that's good 19:26:20 But I'm +1 to us not caring 19:26:28 i dont know what KiB is 19:26:31 that's 4 for not caring 19:26:31 mjg59: we encourage all projects to come to their own opinion and stick with it? 19:26:31 look at https://fedoraproject.org/wiki/QA:Testcase_Mediakit_ISO_Size 19:26:43 It's a bug if something is saying "MB" when it means MiB. 19:26:44 dgilmore: I think it has to do with the oracle of bacon and the presence of kevin bacon in a movie 19:26:46 try writing that without the proper distinction between IEC and SI 19:26:52 Kevin in Bacon 19:26:56 skvidal: likely 19:27:04 It didn't matter in the past because it was just a few bytes more or less. 19:27:05 Kevin_Kofler: The entire storage industry is buggy, then 19:27:10 Kevin_Kofler: as an upstream maintainer, there are several projects which I will not change to meet your bullshit rewording. 19:27:13 Kevin_Kofler: And it makes sense for us to be bug-for-bug compatible with them 19:27:14 But with GiB vs. GB, it makes a HUGE difference. 19:27:20 mjg59: well, I agree with that statement 19:27:39 pjones: Then your projects are broken and must be patched in Fedora. 19:27:50 But really, you ought to fix this upstream too. 19:27:51 Kevin_Kofler: good luck with that. 19:28:06 dgilmore: you +1 on not caring? 19:28:16 -1 to not caring. 19:28:38 skvidal: i dont care 19:28:45 you don't care if you don't care? 19:28:49 We should enforce the KDE policy (or the Ubuntu one which is fairly close AFAICT). 19:28:52 b/c that's SUPER apathy! :) 19:28:56 ok, so that would be 5 then... 19:29:17 +1 on not caring 19:29:32 It's the job of a distribution to be consistent. 19:29:43 #agreed Fesco currently would rather upstream projects form their own policies than anything being enforced in Fedora at this time. 19:29:44 skvidal: im in complete apathy 19:29:49 It's the job of a distribution to be sane 19:29:50 I don't see why we should tolerate apps which don't follow established IEC standards and are inconsistent with what KDE shows. 19:30:06 if i were to recommend anything it would be for banning MiB and freinds 19:30:14 you can only wathc the movie so many times 19:30:15 Kevin_Kofler: You want us to mandate a standard that's incompatible with the entier storage industry? 19:30:25 It's NOT incompatible! 19:30:30 Quite the opposite. 19:30:32 dgilmore: and MiB 2 - they had neat guns - but it was worth it 19:30:40 Kevin_Kofler: Yes it is. Vendors label in MB but mean MiB 19:30:45 pretty sure we're done here. 19:30:52 * nirik personally doesn't mind the MiB junk, but lets move on... 19:30:53 It will stop us from claiming your 500 GB HDD only has some 400 + odds GB. 19:30:54 No, wait. I'm wrong. 19:30:55 Ignore me. 19:31:05 #topic #360 Adjust FESCo Election Policy to match practices of combined elections. 19:31:06 Instead it'll say 500 GB or 4xx GiB. 19:31:12 .fesco 360 19:31:16 nirik: #360 (Adjust FESCo Election Policy to match practices of combined elections.) - FESCo - Trac - https://fedorahosted.org/fesco/ticket/360 19:31:33 +1 for adjusting 19:31:41 its what we have been doing anyway 19:31:45 Yeah, this seems like a no brainer 19:31:49 +1 to adjusting 19:31:51 approve/ 19:32:06 yeah 19:32:20 +1 here, I think adjusting is fine... 19:32:26 +1 19:32:40 #agreed This Adjustment is approved 19:32:55 #topic #362: Incomplete Fedora 13 Features 19:33:14 we have a bunch of these... will go through each in turn I guess? 19:33:29 yep 19:33:37 Well, gnome 2.30 can't be marked 100% until gnome 2.30 is released... 19:33:41 #topic https://fedoraproject.org/wiki/Features/DesktopLiveImageTarget 19:34:02 We approved this, we just need to get them to mark it 100% and finish cleaning it up. 19:34:35 -1 to dropping, we just had this filed, it'd be completely silly for us to drop it. ;-) 19:34:51 Are there any of these features that we want to drop? 19:35:16 multipath might be reasonable to move to f14 (well, to move parts of it there), but I'd just as soon wait on that decision. 19:35:28 well, 'want to drop' I would say no... but there might be some that just don't seem to be done/ready. 19:35:34 (in that I might be able to knock the whole thing out in a day or two if I can get a damn chance ;) 19:35:36 (I'm here, happy to talk about btrfs whenever convenient.) 19:35:38 shall we just backup and look at specific ones? 19:35:57 .fesco 362 19:35:58 nirik: #362 (Incomplete Fedora 13 Features) - FESCo - Trac - https://fedorahosted.org/fesco/ticket/362 19:36:21 It looks like the Python one has hit a snag and is being worked on 19:36:26 Beta is in a week. 19:36:42 Things not 100% by Beta should be dropped/retargeted for next release. 19:37:45 IMHO, the only point at which we can really say that a feature didn't make it is when the last RC for the final release (the one which gets declared gold) is composed. 19:37:53 * nirik has asked mclasen to drop in and help give us status on things. 19:38:09 hey mclasen 19:38:13 hey 19:38:38 I have been negligent of feature page duty... 19:38:44 to many distractions this cycle... 19:38:54 no worries. :) 19:39:03 .fesco 362 19:39:03 nirik: #362 (Incomplete Fedora 13 Features) - FESCo - Trac - https://fedorahosted.org/fesco/ticket/362 19:39:11 this ticket has the list of not 100% ones. 19:39:29 #topic #362: Incomplete Fedora 13 Features 19:39:55 How about we see if there are any that we should drop? does anyone see features that look like they are in danger of not making it? 19:40:17 cjb: Is the btrfs rollback support realistically going to make it? 19:40:55 So, we aimed for three features: kernel support for listing/setting snapshots, a yum plugin for creating new snapshots automatically, and a GUI for choosing which snapshot to boot into. 19:41:03 btrfs is the only one that looks marginal to me. 19:41:20 the intellij-idea one says its 100% 19:41:25 The first two are 100%, and the GUI isn't ready yet because the others took too long. Josef recommends just setting the feature as 100% and deferring the GUI to next release, which I'm inclined to agree with. 19:41:32 (we still have a text UI, just not a GUI) 19:41:42 Yeah, that makes sense. 19:41:42 dgilmore: I think we've got about a week of lag between making that list and looking at it, so that's not surprising. 19:41:44 cjb: Ok 19:41:47 cjb: sounds reasonable. 19:41:52 cjb: that sounds fine 19:41:52 nirik: gnome 2.30 is 99% done, I need to do some 10 or so builds to get everything in 19:41:52 cool, will do that, then. thakns. 19:42:00 mclasen: The user admin dialog? 19:42:06 mclasen: yeah, saw the 2.30 builds landing. ;) Cool. 19:42:21 user accounts dialog is also basically done for f13 19:42:28 mclasen: Ok, sounds good 19:42:28 halfline built the gdm part last week 19:42:37 and I landed a big code review of the service and dialog today 19:42:52 I suspect virtx2apic just needs a link to the documentation 19:42:59 So I don't think we need to drop anything right now 19:43:14 I see 2 features not updated recently and not at 100%: 19:43:20 desktop live image size is largely walters' child at this point, and still a little fuzzy 19:43:21 * nirik looks at moblin and sugar 19:43:21 Virtx2apic and YumLangpackPlugin 19:43:23 moblin 2.2 i think we should wait and see what the fall out is 19:43:35 Both are about 2 months out of date and the status doesn't contain any details. 19:43:36 * pjones steps out to use the restroom 19:44:00 * mclasen goes to update some of the feature pages 19:44:18 I'm also a bit worried about the Moblin status as upstream is renaming the final release, so it looks like we'll have tons of rename reviews to deal with. 19:44:50 The remaining features all look like they'll definitely make it. 19:45:26 * pjones returns 19:46:16 I can bother the YumLangPackPlugin feature owner later tonight... 19:47:54 I note that our next meeting is after Beta is out. 19:48:10 so, shall we leave all these, and do another pass later in the week? or monday? 19:49:00 thoughts? 19:49:19 do we have more to go through other than these? 19:49:21 i think we should push for them to be updated and finished 19:49:33 skvidal: i think these are it 19:49:38 the resta are all done 19:49:46 skvidal: for features? just these... the rest are 100% 19:49:58 nirik: no - I mean non-feature items for the meeting 19:50:02 do we have anything else to cover? 19:50:04 i'm fine with the list as it is (YumLangPack followup would be nice) 19:50:08 oh 19:50:19 oh, sorry... just a FES update... and open floor. 19:50:42 I think we shouldn't drop any of those features at this time. 19:51:09 dgilmore: yeah, we should press them to update for sure... but should we drop anything now? 19:51:32 nirik: i dont think so. 19:51:42 But we should nag folks to provide status updates and to get things to 100% ASAP. 19:52:06 It would be nice to be able to check before Beta goes out. 19:52:09 (and if the stuff which is written there is not 100%, then they need to change the feature page to mention only what's done, as in the process) 19:52:41 nirik: I don't see how that's relevant. 19:53:11 So far it looks like all the features are done or almost done in the beta and will be done in the final release. 19:53:18 Kevin_Kofler: because we are going to advertise these with the Beta release... and if they don't get finished and must be dropped, we should really not do that. 19:53:47 Well, I don't see any of the features there which is so incomplete it cannot be advertised. 19:53:55 yeah, probibly true. 19:54:14 ok, I guess we all try and get things updated and just move on? Any further discussion on this topic? 19:54:24 Not from me 19:54:38 nope. 19:54:43 #topic FES update 19:54:50 https://fedorahosted.org/fedora-engineering-services/report/6 19:54:52 I do think the Yum Langpack Plugins and Virtx2apic features could use updating, and if not 100%, a more detailed report of what's missing (and possibly refocusing the feature). 19:55:09 Nothing big here... some new unassigned tickets for interested folks tho. 19:55:17 https://fedorahosted.org/fedora-engineering-services/ticket/15 19:55:29 port syslinux isohybrid perl script to C 19:55:45 otherwise tickets moving along... deps being fixed, etc. 19:56:15 mmcgrath: anything to note from you this week on FES? 19:56:44 nirik: nope, we did a lot of work to determine that there's only a couple of pacakges causing issues at the moment. 19:56:55 which is good I suppose, once they're fixed rawhide should be in good shape again.. 19:57:06 excellent. 19:57:10 We got a few more volunteers to join up and they all have work at the moment, I have a couple more to look through. 19:57:27 #topic Open Floor 19:57:34 Anything for open floor? 19:57:38 FYI: https://fedorahosted.org/fedora-engineering-services/ticket/15#comment:1 19:58:40 yeah, xfconf also has a perl binding (although we might be able to split it out and not ship it by default in the spin) 19:59:10 ok, if nothing else, will close the meeting here in a minute. 19:59:53 ok, thanks for coming everyone! 20:00:07 thanks nirik 20:00:09 tt 20:00:11 #endmeeting