18:01:19 #startmeeting FESCO (2013-10-30) 18:01:19 Meeting started Wed Oct 30 18:01:19 2013 UTC. The chair is notting. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:01:19 Useful Commands: #action #agreed #halp #info #idea #link #topic. 18:01:22 Hello 18:01:26 #meetingname fesco 18:01:26 The meeting name has been set to 'fesco' 18:01:26 #chair abadger1999 mattdm mitr mmaslano notting nirik pjones t8m sgallagh 18:01:26 #topic init process 18:01:26 Current chairs: abadger1999 mattdm mitr mmaslano nirik notting pjones sgallagh t8m 18:01:28 hey. 18:01:35 * nirik runs to get coffee, back in a minute 18:01:46 * mattdm has coffee in hand 18:02:11 I'm around; closing out the Server WG meeting 18:02:31 I'll be leaving in 55 minutes to start the cloud wg meeting :) 18:03:19 hello. 18:03:33 * abadger1999 shows up 18:03:42 mattdm: sgallagh: so this is becoming an increasingly bad time slot then, eh? 18:04:05 pjones: It works for me; I like having an upper limit on the Server WG slot 18:04:10 pjones or we could keep this meeting to an hour :) 18:04:19 is that everyone except mmaslano? 18:04:27 looks like it 18:04:30 but for the cloud wg, this week probably isn't representative 18:04:46 (not sure if we will even be able to find a regular time) 18:05:21 hi 18:05:43 also, US changes time next week... 18:05:44 ok, i believe that's everyone. 18:05:45 Before we get started. 18:05:52 nirik: beat me to it 18:06:33 (and eu changed already) 18:06:43 is everyone ok with the meeting staying at 1800UTC , with their local time shifted back? 18:06:51 I am ok with that 18:06:57 actually I prefer that 18:07:13 sure. 18:07:17 * nirik hates timezones 18:07:44 i slightly prefer the later time but either works 18:07:48 Works for me. 18:07:50 yeah, it's okay 18:07:56 * t8m hates DST 18:08:01 So that moves it to 1:00 EDT next week? 18:08:05 err, EST 18:08:11 if we stay at 18UTC, we might ask the server and cloud WG's to also stay if they don't want to overlap 18:08:23 sgallagh: right 18:08:29 t8m: don't we all 18:08:36 In the case of the Server WG, we're going to do another WhenIsGood in any case 18:08:42 Ditto 18:08:45 pjones, sure :) 18:09:01 (with the distinct possibility of NothingIsGood) 18:09:15 mmaslano: speaking of WG meetings, we haven't arranged an Env and Stacks meeting yet, have we? 18:09:30 abadger1999: Working on it 18:09:31 workstation wg is going to just avoid live meetings for the time being 18:09:39 proposal: keep meeting at 1800UTC until further notice 18:09:43 abadger1999: just writing email 18:09:43 Cool. Just making sure I didn't miss any announcements. 18:09:43 +1 18:09:48 notting, +1 18:10:04 +1 18:10:05 notting: +1 18:10:08 +1 18:10:19 +1 18:10:48 +1 18:11:09 +! 18:11:21 #agreed meeting stays at 1800UTC until further notice (+:9) 18:11:43 ok, ye olde business 18:11:47 #topic #1170 Working Group call for volunteers 18:11:47 .fesco 1170 18:11:48 notting: #1170 (Working Group call for Volunteers) – FESCo - https://fedorahosted.org/fesco/ticket/1170 18:11:49 that actually makes my wednesday schedule /way/ easier. 18:12:16 just as long as meetings don't still go to 2030UTC :) 18:12:34 ok, on this we need to approve the base design group? 18:12:35 I think we can close the call-for-volunteers ticket 18:12:40 oh wait that. :) 18:13:00 Ok, as noted on the mailing list, jwb isn't insistent on his membership and I'd kind of like to see Lennart on board here. 18:13:26 well, what does pknirsch think? 18:13:32 unknown 18:13:52 i'd be fine with that 18:14:00 ah, good. pknirsch is here 18:14:05 ofc :) 18:14:20 he must have his own reason for not choosing him and bunch of other people one might have expected there 18:14:20 sorry, was just lurking and doing other stuff on the side 18:14:28 the idea was to have either Harald or Lennart on the group 18:14:31 With the "This is the common base on which the Fedora OS products will be built. " wording it's not clear to me that systemd is a core aspect - I'd rather have the toolchain and basic libraries as the fox (I know others have a different opinion) 18:14:48 er 18:14:50 s/fox/focus/ 18:15:06 i fail to see how the init process isn't part of the core aspect of the OS, particularly given it's current place 18:15:20 and the role it will play with cgroups going forward 18:15:24 As well as its ever-increasing roles 18:15:29 yeah, I'm pretty sure we need systemd there 18:15:43 I don't really see any way systemd wouldn't be part of it, but that's not what's important here. 18:15:43 well, harald is already there too... 18:15:49 jwb: Booting a localized implementation detail, same as KMS for display is. (cgropus are another matter) 18:15:55 or as lennart calls it: "Das PID Eins" hihiihih 18:15:57 nirik: 18:15:57 mitr, +1 18:15:58 the question is: do we have people familiar with the sort of problems the group will face 18:16:10 but note also that this is just voting members, I surely expect involved people to be... well, involved. 18:16:21 +1 pjones, also +1 nirik 18:16:23 mitr, if that's your perspective, i think you've clearly missed what is going on in fedora for the past 5 releases 18:16:41 systemd has increased scope. ignoring that as "implemenation detail" is disingenuous 18:16:58 anyhow, pknirsch: would you like to amend the proposal base group any? or leave it as mentioned in the ticket? 18:17:13 jwb: In my view, the APIs that are a part of the increased scope are important - the fact that it is tied to systemd isn't 18:17:17 s/has increased/continues to increase/ ;) 18:17:42 Anyway, I'll not be voting if there is a vote. 18:17:57 Proposal: Substitute Lennart for Josh, accept the rest as-is? 18:18:03 mitr, sure. and the APIs amazon provides are important to cloud development. but amazon is just an implemenation detail. 18:18:10 mitr: I think that may be a fine long-term distinction. short term, they're probably equivalent (we're still i nthe stage where we are growing the apis along with a single implementation). 18:18:12 personally, I would have liked to se lennart on this list, but I do think the current list is pretty good. 18:18:14 sgallagh: would prefer that proposal come from pknirsch 18:18:17 sgallagh: -1 18:18:19 jwb: "Booting" is an implementation detail, not the APIs 18:18:21 what notting said. 18:18:27 * jwb just shuts up 18:18:43 sgallagh: 0 18:19:07 pknirsch: so, what say ye? 18:19:37 well, whats the reason against harald? I know lennart is leading the systemd development, but kay is doing udev and harald is dracut, all of which would be essential, too 18:19:41 same goes for anaconda 18:19:44 i guess i would say i, as fesco member ,would approve the list both as is, or with s/jwb/lennart/. 18:20:01 pknirsch: No one said anything against (well, anyone). 18:20:04 yeah, me too. 18:20:34 don't get me wrong, i have nothing against lennart, but harald can cover the topics just as well imho 18:20:58 pknirsch: don't you miss someone from security? 18:20:59 I'm -1 as-is, but not because I don't like jwb :) I just think that the guy who's doing all of the work should have a say. 18:21:14 mmaslano: Dan walsh 18:21:18 it's not clear to me which area will be covered by Base, so it's hard to say, whoshould be there 18:21:21 * nirik does feel a little annoyed at lennarts reaction to not being chosen in the first place, but perhaps thats just me. 18:21:23 okay 18:21:33 so especially sgallagh, what are your reasons for wanting lennart on the team 18:21:35 ? 18:21:41 sgallagh: hopefully nobody is doing /all/ the work. and as pknirsch rightly pointed out, there's more than just systemd involved. 18:21:44 * abadger1999 would like to see jwb on there as the kernel guy 18:21:58 sgallagh: also hopefully skilled, involved people will have /a say/ even if they don't get /a vote/. 18:22:06 abadger1999, i'll still do the kernel thing either way. my involvement doesn't hinge on voting 18:22:12 18:22:12 * pknirsch nods 18:22:13 so i'm good with whatever 18:22:17 if you don't get to make salient points and influence others because you're not a voting member, something has gone horribly wrong. 18:22:34 my expectation is that anyone who wants to get involved should just do so 18:22:53 and come to the meetings, voice his opinions and discuss with us what the product will look like 18:23:10 and everyone is invited to do so 18:23:17 +∞ 18:23:28 Shall we just vote on the proposal as-is? If I'm the only hold-out, so be it. 18:23:50 And yes, I agree that non-voting members must still be valued participants 18:24:05 i'm +1 to the proposal in the ticket. and/or abstaining b/c i'm on it. 18:24:09 +1 as it is. 18:24:13 +1 as it is 18:24:19 +1 as it is 18:24:21 I'm +1 to it. 18:24:31 +1 as is 18:24:35 +1 as well. 18:24:49 +1 to having something / 0 on the specific membership question 18:25:18 I do hope that as this develops we are able to bring more non-rh community members in at this level. 18:25:30 ++ 18:26:10 Seems decided then. Let's move on. 18:26:26 all done with this ticket and close? 18:26:38 Yes 18:26:50 #agreed base design WG approved (+:8, -:0) 18:27:02 perhaps open a new one to trac gov docs proposals (due nov 15th) 18:27:28 nirik, +1 18:27:29 ok. ticket #1180 was closed out after being added to the agenda (yay!) 18:27:38 small request: as we're a bit behind for Base Design, this might be delayed by a week (potentially), but we'll see 18:27:42 notting: That should have been -1, sorry if it wasn't clear 18:27:48 other groups had a head start! 18:27:48 :) 18:27:50 A weak -1 18:27:52 #undo 18:27:52 Removing item from minutes: 18:27:56 #agreed base design WG approved (+:8, -:1) 18:28:17 nirik: would you like to open said ticket? 18:28:41 sure, I can... 18:29:13 pknirsch: let's ask them to meet and see if they really need it pushed back first? 18:29:17 #info nirik to open a ticket to track working group governance proposals 18:29:17 it's not a race after all ;) 18:29:38 pjones: But if it was, you'd be winning? :) 18:29:40 I'd be happiest if one group finished the charter and the others just copied it unmodified :) 18:29:43 new business. 30 minutes or less! 18:29:50 #topic #1185 Enable "-Werror=format-security" by default 18:29:51 .fesco 1185 18:29:54 notting: #1185 (Enable "-Werror=format-security" by default) – FESCo - https://fedorahosted.org/fesco/ticket/1185 18:30:20 mitr, yep that would be good :) 18:30:20 i'd agree this seems reasonable, but i'm willing to wait on mass rebuild results to see how doomed we are 18:30:24 I'm +1 to this, really, though I would be interested in the mass rebuild results 18:30:29 Proposal to defer this until we see how bad it is? 18:30:35 sgallagh: +1 18:30:42 sgallagh: sure, +1 18:30:55 sgallagh, +1 18:31:06 sgallagh: yeah, +1 18:31:11 I mean, I strongly advise all upstreams to voluntarily enable that flag, but who knows how many packages are already out there that will fail. 18:31:12 sure, +1 18:31:37 +1 18:31:38 +1 18:32:40 #agreed defer pending results of mass rebuild (+;8, -0, 0:0) 18:32:53 #info packagers are encouraged to voluntarily enable and test for issues 18:33:01 sgallagh, halfie: The logical conclusion would be to enable the flag in the compiler by default (... at least as a warning) 18:33:23 +1 18:33:26 #undo 18:33:26 Removing item from minutes: 18:33:28 (not a FESCo matter at this point) 18:33:30 #agreed defer pending results of mass rebuild (+;9, -0, 0:0) 18:33:36 We should probably ask for one example of fixing code. 18:33:44 sgallagh: well, there's still the question of how reliable the detection is, but yeah 18:34:02 mitr: The warning is already enabled as part of -Wall these days, I think 18:34:02 (in terms of false negatives or false positives) 18:34:04 abadger1999: good idea 18:34:45 there are links in the ticket that show you what to do, i believe 18:34:55 pjones: It's pretty simplistic: if the format specification is anything but a static string, it throws a warning 18:35:00 (in short, printf(buf) -> printf("%s", buf) ) 18:35:00 sgallagh: it's not even in -Wextra on F19 18:35:29 mitr: Really? That *is* wrong. Ok, yeah. Let's ask for -Wformat-security added at least. 18:35:29 well, then it's also very easy to fix then 18:35:47 halfie: would you care to file a gcc RFE? 18:35:54 mitr: gcc? 18:35:59 Why isn't it in -Wall ? 18:36:14 sgallagh, it should be in -Wall 18:36:14 proposal: we ask gcc to make this warning part of -Wall now, and defer for enabling it as an error until the mass rebuild is done 18:36:15 I was just thinking redhat-rpm-config, but adding it to -Wall seems smart too 18:36:23 pjones, +1 18:36:38 pjones: +1 18:36:45 pjones +1 18:36:46 sgallagh: redhat-rpm-config fixes the distribution; gcc has a chance of fixing the ecosystem 18:36:48 pjones: +1 18:36:52 pjones: +1 18:36:56 * pjones also +1 18:36:57 +1 18:37:00 mitr: Yeah, I wasn't thinking big enough. You are absolutely right. 18:38:00 sure, although there might be reason for it. 18:38:22 nirik: Then that will be communicated back in the bug report. 18:38:28 #agreed ask gcc to make warning part of -Wall now, defer enabling it as an error until the mass rebuild test is done (+:7, -:0, 0:0) 18:39:20 halfie: can you take care of that RFE? 18:40:22 ... ok, will put that in the ticket. 18:40:24 moving on 18:40:31 #topic #1186 FESCo liason role in WGs 18:40:31 .fesco 1186 18:40:33 notting: #1186 (FESCo liason role in WGs) – FESCo - https://fedorahosted.org/fesco/ticket/1186 18:41:07 so there are really 2 questions here, if you want me to summarize 18:41:13 please do :) 18:41:14 Proposal: revamp FESCo selection process so that five of the seats are populated from the WGs. 18:41:25 holy balls 18:41:27 * nirik waits for jwb 18:41:36 sgallagh, i think that's another ticket entirely. 18:42:01 so the two questions on the liaison role are this: 18:42:22 1) is the liaison always supposed to be a voting member of the WGs 18:42:28 sgallagh: that's kind of terrible 18:42:44 2) is the liaison always appointed by FESCo, or can that role be filled by someone the WG chooses 18:43:03 sgallagh: Let's let the board revamp occur first ;-) 18:43:24 jwb: I would think 1) yes, and 2) the former 18:43:26 jwb: Right, and my counter-argument is that the WG could simply volunteer one of its voting members to be their seat on FESCo. 18:43:39 #1 I would agree with mattdm and say yes. 18:43:57 jwb: in order, 1) yes 2) no 18:43:58 sgallagh: but since fesco approved who's on the WGs, at that point fesco has elected itself. 18:43:59 I agree with myself, ftr 18:44:00 * nirik isn't sure fesco can redo how it's formed, that might be a board question/ack? 18:44:13 nirik: I'm not sure we want to anyway 18:44:17 notting, 2) no for what? 18:44:24 pjones: Not after the governance is decided. 18:44:40 sgallagh: you may be just a *bit* ahead of yourself 18:44:41 #2 I lean towards it being a role like "secretary" which can be determined by the WG (and vary from week to week) but I could easily be swayed. 18:44:42 t8m: sorry. 2) no, not always appointed by fesco. up to WG. 18:44:50 1) not essential, but "yes" is more practical, 2) yes - let's make this (formally) one-directional instead of interdependent; we can always talk to each other 18:45:22 so it looks like we've got some differing opinions about question #2 18:45:26 ok, so far everyone says the liaison is required to have a voting seat. can you settle that one and then get to question 2 since none of you agree 18:45:27 1) yes, 2) meh... could be swayed either way there. 18:45:30 mitr: I agree with mitr 18:45:44 1) yes 2) undecided 18:45:45 proposal: the liaison role is always a voting seat in the WGs 18:45:46 Proposal: require that the fesco liason on a WG always has a voting seat 18:45:50 nirik: Well, fesco originally formed itself without Board governanance -- but likely at this point we'd ask the Board to confirm changes that we come up with -- weI think this tangent is in the weeds though. 18:45:55 jwb/pjones: +1 18:45:57 jwb: you don't get to make proposals in this meeting any more do you? ;) 18:46:00 * pjones +1 to himself too 18:46:02 pjones: +1 18:46:03 jwb/pjones: +1 18:46:08 pjones, i do not. got ahead of myself. vote on pjones 18:46:16 why on earth cant the groups a) contact fesco if needed on their own and b) appoint one additional willing participating member other then fesco ? 18:46:17 +1 18:46:18 pjones, +1 18:46:18 #1: +1 18:46:20 um, one monkey wrench. what if the WG decries voting as its decision making mechanism? 18:46:34 heh :-) 18:46:35 notting: well, that'll be odd. 18:46:35 notting, i dislike you. 18:46:48 but I think we just inadvertently voted to ban that, too. 18:46:49 notting: I doubt we'd approve that governance 18:46:56 in that case, i would assume 1 turns into "the liaison role must exist within the WG" 18:46:58 notting eh. "voting" = "member of core group whatever that is" 18:47:12 Viking-Ice: My thinking is that the voting seat implies an obligation to be present and involved => ensures that the person is actually actively participating in both FESCo and the WG and fully aware of all agenda items 18:47:13 if you want to phrase it as 'member of the WG's decision-making body', that's fine. 18:47:14 +1 jwb 18:47:19 +1 notting 18:47:21 sure, that's fine. 18:47:22 +1s all around 18:47:24 notting: I'm good with that wording 18:47:27 mitr, but still not necessary 18:47:47 yeah sure for the initial members but tying fesco to participate in this forever 18:48:08 notting, turn that into a counter-proposal to vote on? 18:48:10 tying everything together is exactly the point 18:48:28 #agreed require that the fesco liason on a WG is always a member of that WG's decision making body (+:8, 0:0, -:0 18:48:35 or that 18:48:35 pjones: yeah, +1 to that 18:48:38 the parens will just have to stay broken 18:48:47 notting: oh, I counted your verbal affirmation as a yes 18:49:04 FESCo's purpose is to direct technical decisions in Fedora. If we aren't directly involved in the WGs, we're redundant. 18:49:05 * pjones recounts 18:49:06 and what happens if that liason gets outvoted 18:49:16 fesco steps in and overtakes the whole thing 18:49:17 yeah, it's right. 18:49:20 Viking-Ice you mean within the group? 18:49:28 Viking-Ice: then the liason gets out voted. 18:49:31 Viking-Ice: nothing happens 18:49:38 Within reason... 18:49:40 by default, the vote is the vote. 18:49:48 not nothing. then it's their responsibility to communicate to fesco the vote of the group 18:49:49 then they are... unless it's something so dire it's brought back to fesco to overrule I guess. 18:49:50 the point is to have a defined point of contact for fesco to get decisions that are the Voice of the WG. if the liason is misrepresenting his personal opinions as the will of the WG, that is a separate issue that can be solved. 18:50:02 people can bring any decisions up with fesco no matter how fesco's liason votes. that doesn't mean we'll vote to reverse, and it doesn't mean we won't. 18:50:05 +1 notting 18:50:19 Viking-Ice: then they have to report to fesco the desires of the wg -- they can give their minority opinion as well.. but then, any member of the WG can do that as well. 18:50:19 right, so on to the second part? 18:50:35 please 18:50:49 as i said in the ticket, I don't think it's _necessary_ 18:50:57 I'm not sure it's actually reasonable to answer #2 right now? 18:50:57 Proposal: liason is always assigned by FESCo 18:50:59 I'm -1 18:51:09 +1 18:51:09 I mean, it depends on what kind of governance the WGs want, really? 18:51:36 thus far that has been "minimal" 18:51:39 What if we phrased it "the liason is always approved by fesco"? 18:51:54 pjones, +1 18:51:55 thus letting us let either group pick it as long as fesco is okay with who gets picked. 18:52:01 there is a difference in liason be assigned to fesco by the wg themselves or fesco assigning a liason to the wg 18:52:05 pjones: I'm good with that. 18:52:08 confirmation hearings? ;-) 18:52:16 I'm +0. I'm having trouble imaginging a plausible scenrio where it would really be a problem yet there isn't some even more gigantic problem which fesco picking a liaison would not solve 18:52:27 pjones: works for m 18:52:29 abadger1999: we could have a grandstanding competition at flock ;) 18:52:30 that could probably use some punctuation 18:52:30 pjones: i'd even say that's overkill 18:52:51 notting: just give us the right to reject one, like an ambassador? 18:53:10 whichever poor soul a WG nominates to take crap from both sides... ok. 18:53:18 if we reject someone and appoint another one, that bumps a member from that group right? 18:53:26 I'm kind of coming down on -1 here 18:54:01 if a wg nominates someone and we somehow can't get along, we have a bigger problem 18:54:03 Honestly I'm really not sure we need to have a firm rule on this 18:54:21 well, WG's need to know right? 18:54:33 if we are appointing then they only have 8 seats to fill... 18:54:35 Clearly we have the right to remove somebody from that roll, since we're the overseeing body above the WGs 18:54:36 nirik, I would think if fesco ( for whatever unimaginable reason ) would reject the wg assigned as an acting liason between fesco and the wg it would only grant fesco the rights to choose from other voting members of that wg 18:54:52 yeah, i believe jwb wanted to know so he can clarify his governing docs 18:54:53 Viking-Ice: yeah, could be. 18:54:58 nirik: actually -- WGs are allowed to determine thenumber of people on them as well. 18:55:01 i would like to know, yes 18:55:15 nirik: but yeah, they may want to plan an extra seat because of that. 18:55:35 abadger1999, after Jan i believe that is true 18:55:44 until then, it's 9 and the 9 approved by fesco 18:55:56 jwb: Yeah. 18:56:24 proposal: WGs can decide how the fesco liaison is selected, including the possibility of asking fesco to make the selection 18:56:41 mattdm, I can be +1 to that 18:56:42 I'm kinda with pjones -- if we can't get along with the liason from the WG our first step would probably to let the WG know that there's an issue with the person representing them to us. 18:56:46 mattdm: +1 18:57:18 at which point the rational next step is simply for them to select someone else on the WG to liase with us. 18:57:21 mattdm: I can get behind that. Obviously still with the caveat that if they chose to chose, we can always ask them to chose again. 18:57:31 mattdm: -1; FESCo should have the ultimate authority 18:57:40 sure, +1 to mattdm's proposal. 18:57:48 mitr: I don't think we can even /give away/ the ultimate authority. 18:57:59 mattdm + pjones caveat: +1 18:58:04 * mattdm has 3 minutes 18:58:11 mattdm: +1 that seems reasonable 18:58:13 is fine with caveat if it comes down to that 18:58:27 mitr: FESCo retains the right to disband the WGs at any time, if it came to that 18:58:34 I don't think it gets much more "ultimate" 18:58:37 * pjones is +1 to mattdm+caveat that uses chose as much as possible 18:58:46 sgallagh: with that, is the caveat even necessary? 18:59:02 notting: no, it's really implicit. hence "obviously" 18:59:05 * abadger1999 notes that two of those "chose"s should be "choose" 18:59:11 abadger1999: yes ;) 18:59:23 abadger1999: or arguably only the last one, but still. 18:59:30 * mattdm is outta here. 18:59:53 * frankieonuonga is here. 19:00:04 * frankieonuonga waves 19:00:42 i didn't see a #agreed 19:00:52 frankieonuonga: #fedora-meeting-1 for Cloud WG 19:00:52 I'm not sure anything was agreed upon 19:01:02 There are 5 +1's -- did this pass or do we want to clarify whether we're leaving the caveat implicit or explicit? 19:01:07 I think we have +1s 19:01:16 #agreed WGs can decide how the FESCo liason is selected, including the possibility of asking FESCo to select. (As FESCo is above the WGs, FESCo could ask WGs to re-choose.) (+:7, -:1, 0:0) 19:01:25 sorry, was counting 19:01:25 that works 19:01:37 pjones: didn't mean to materially re-phrase 19:01:48 looks fine to me 19:02:02 jwb: does that cover your concerns? 19:02:03 ok, thanks for the clarifications. i now return you sgallagh wanting to completely restructure your group ;) 19:02:16 * sgallagh is willing to leave that for another day 19:02:26 ok. sgallagh - please file a ticket if you have a proposal 19:02:43 sgallagh: thanks 19:02:47 lastly (there were two late tickets we can leave for next week).... 19:02:47 notting: With the parentesized part, I can be +1 (not that it matters) 19:02:56 #undo 19:02:56 Removing item from minutes: 19:03:03 #agreed WGs can decide how the FESCo liason is selected, including the possibility of asking FESCo to select. (As FESCo is above the WGs, FESCo could ask WGs to re-choose.) (+:8, -:0, 0:0) 19:03:13 mitr: nice typo. 19:03:20 #topic #1187 Packagers should be not be allowed to ignore RH bugzilla 19:03:20 .fesco 1187 19:03:21 notting: #1187 (Packagers should be not be allowed to ignore RH bugzilla) – FESCo - https://fedorahosted.org/fesco/ticket/1187 19:04:58 There's not enough info here really to do much on IMHO 19:05:06 I think the initial bug report is just pointing out a problem. 19:05:14 handsome_pirate: you around? 19:05:30 so... i don't think we can mandate a particular level of responsiveness to individual bugs other than what's clarified by the non-responsive mantainer policy, but i do think "blocker bugs are filed in RH bugzilla; ignoring them is not appropriate maintainer behavior" is reasonable 19:05:32 I agree with the general sentiment, but our options are limited 19:05:47 There are a lot of package maintainers who don't pay attention to bugs and just ship out upstream releases 19:06:20 mitr had one partial solution. another option is removing people from packager group or orphaning/retiring packages that accumulate too many (of some class?) of bug. 19:06:21 notting, +1 19:06:26 and there's also maintainers who have priorities that may not match all the people who file bugs against their packages priorities. 19:06:31 Yeah, I think we should just say that people who ignore blocker bugs shortcut the non-responsive policy 19:06:51 mitr: I vaguely like your suggestion, but possibly not /any other sponsored packager/ - there's no reason to encourage personality conflicts 19:06:54 I'd be curious what bug(s) prompted this? were they blockers? proposed as blockers? 19:07:00 mitr: at the same time - isn't this why we have proven packager? 19:07:03 nirik: for example, for better or worse, individual abrt bugs that aren't heavily duplicated are waaaaaaaay at the bottom of my priority 19:07:17 notting: me too 19:08:09 but for example, there was someone in #fedora the other day very intent on the vty bug that doesn't display the f19 release right... but I can see that not being very high on the kernel maintainers to do anything with. ;) 19:08:29 nirik: closed->get a better vga bios font 19:08:36 yeah 19:08:55 anyway... 19:08:58 the kernel is the worst case bugzilla example ever. pick something else please 19:09:00 if you want this fix open up contribution for everybody to fix spec files without having a hard requirement on proven packager or being a packager 19:09:37 pjones: provenpackagers are busy - I wanted to give interested newbies access to core packages (= motivation for newbies to look at core packages more, and motivation for owners to core packages to work on them more to keep them if they don't want newbies). The personality conflict aspect is therefore kind of intentional - which may very well be a horrible idea. 19:09:41 that will solve at least the packaging bbug 19:09:43 sgallagh: hm. i guess one issue is that anything that has a blocker is likely something that shouldn't be shortcutted to orphaning/blocking, which means we'd need a maintainer, or provenpackager shephard, ayway 19:10:05 Viking-Ice: that also opens up to a flood of bugs/brokenness from people who don't understand the package... 19:10:24 Or security issues from people who understand it too well... 19:10:43 nirik, if we cant a) detect that b) revert that then we have more serious issue 19:10:43 or people ignoring freezes, etc. 19:11:04 Viking-Ice: pragmatically we can't detect that, so reverting is moot 19:11:32 mitr: Well, the other official maintainers will get an email that something was changed 19:11:39 So presumably they can take action as needed 19:11:44 isn't "detecting that" the same issue we're talking about here to begin with? 19:11:51 proposal: if blocker bugs are ignored, escalate to fesco. 19:11:57 sgallagh: sure, but they also get bz emails and whatnot 19:12:03 nirik: and what will we do? 19:12:16 for me it would depend on the situation. 19:12:17 * mitr notes two bug escalations have been filed in the past 3 hours 19:12:29 so, what part of 'deal with reported bugs in a timely manner' section on https://fedoraproject.org/wiki/Package_maintainer_responsibilities is incorrect/needs fixing? 19:12:32 proposal: let people come to fesco on a package-by-package basis if they think there are package maintenance issues 19:12:33 I'm still waiting for pp to step in and fixing the cron dependence ( yeah and those had patches ) so people cant act like they are whole sale savior of them all 19:12:34 pjones: Certainly, but I suspect people will pay more attention to "l33thacker3 pushed a commit" than BZ 19:12:39 (i.e. exactly what we're doing now) 19:12:54 pjones: +1 19:12:57 pjones, this is probably just #info because they can surely do that already 19:13:02 sgallagh: if they're dumb enough to use l33thacker3 as a nick, I doubt if that much attention is needed 19:13:06 t8m: sure. 19:13:06 pjones: Well "timely" and "accurately" are different 19:13:20 pjones: hahaha 19:13:30 i'd almost say the wiki part overstates it, because it assigns responsibility to address (all?) filed bugs, even including abrt crud, etc. 19:14:03 t8m: I guess I read an implied "and close this ticket, nothing to do without more info" there. 19:14:29 nirik, +1 19:14:48 perhaps I was reading too much into pjones proposal tho 19:14:52 nirik, btw there exist no place or process for us that want to fix packaging issues as we come across them without having to become first a) packager b) proven packager. maybe people arent interested and maintaining package in the distribution and just want to fix things because it annoy's them 19:15:16 nirik: no, you're reading it right 19:15:22 Viking-Ice: Then they can submit patches 19:15:30 And if the patches are ignored, escalate that to FESCo? 19:15:37 We had a case like that a few months ago 19:15:38 sgallagh, which get fracking stuck in bugzilla 19:15:41 Viking-Ice: there was some talk about making a process for mass changes that was better, but not sure it ever went anywhere. 19:15:46 We ended up removing a maintainer for that reason 19:15:58 sgallagh, did escalate that to fesco 19:16:11 and I still have to deal with 50 more packages then 500 on top of that so... 19:16:52 Viking-Ice: Have you requested provenpackager access? 19:17:12 sgallagh, you seem to have forgotten that you and or abadger1999 did so years ago 19:17:13 sgallagh: I requested provenpackager for Viking-Ice for a specific set of changes. 19:17:41 Viking-Ice: I have trouble remembering what I did this morning, so forgive me. 19:17:49 sgallagh, but due to account of one single maintainer I got thwerted and yes I still think nfs is broken 19:18:11 I think this is a different case, and we are driving off in the weeds. 19:18:16 Viking-Ice: Please file a ticket and we'll deal with it next week 19:18:48 pjones' proposal was: let people come to fesco on a package-by-package basis if they think there are package maintenance issues 19:18:56 * pjones is +1 to that 19:19:00 pjones: +1 19:19:01 * nirik is also +1 to that 19:19:20 * sgallagh oO(isn't that the current state?) 19:19:25 sgallagh: yes 19:19:28 yes. 19:19:29 +1 although this is the current state 19:19:36 +1 it is current state 19:19:41 * notting is +1 19:19:49 The followup from this would be asking handsome_pirate to s/gnome/$specific_packages/ for next week? 19:19:50 +1 19:19:59 mitr: sure. 19:20:04 #agreed let people come to fesco on a package-by-package basis if they think there are package maintenance issues (+:7, -:0, 0:0) 19:20:16 should we re-iterate the bug handling portion of the maintainer responsibilities? 19:20:17 mitr: yes. 19:20:27 Maybe mention the portion of https://fedoraproject.org/wiki/Package_maintainer_responsibilities that notting quoted earlier in the ticket. 19:20:38 notting: Can't hurt 19:20:41 #info if the person filing the ticket has specific issues, he should bring them up, as per: https://fedoraproject.org/wiki/Package_maintainer_responsibilities 19:20:43 notting: Yeah, +1 19:21:10 #info Note that handling bugs is part of package maintainership https://fedoraproject.org/wiki/Package_maintainer_responsibilitiesresponsibilities, per 19:21:22 notting: might have typoed a bit there 19:21:30 #undo 19:21:30 Removing item from minutes: 19:21:42 #info Note that handling bugs is part of package maintainership, per https://fedoraproject.org/wiki/Package_maintainer_responsibilitiesresponsibilities 19:21:46 #undo 19:21:46 Removing item from minutes: 19:21:49 paste-os everywhere 19:22:04 #info Note that handling bugs is part of package maintainership, per https://fedoraproject.org/wiki/Package_maintainer_responsibilities 19:22:36 #topic next week's chair 19:22:40 volunteers? 19:23:02 I haven't been in a while, I'll do it. 19:23:14 #info abadger1999 to chair next week's meeting 19:23:22 #topic Open Floor 19:24:08 anything? i 19:24:10 heh, if I get the UTC/DST transition wrong, someone ping me to start the meeting ;-) 19:24:16 i have one thing 19:24:17 i'll leave the bug escalations for next week's meeting. 19:24:55 jwb: ok 19:24:57 i've seen several references scattered around about waiting for a board restructure for various things. is that something FESCo is actively waiting on? 19:25:12 * nirik hasn't really been 19:25:20 * abadger1999 hasn't 19:25:34 I would wait on a fesco restructure until the board restructure happens. 19:25:59 ok, well, in general that whole thread is on a burner back far enough that it is almost falling off the stove. 19:26:03 so don't wait for it. 19:26:14 :-( 19:26:34 alright 19:26:36 * abadger1999 was looking forward to that 19:26:43 or if you would like to see progress, push for it 19:27:02 where "it" is change, not necessarily what i proposed 19:27:04 * nirik recalls the gov discussion at fudcon tempe... never really went anywhere tho 19:27:26 i was not present 19:27:47 i will say that any such change is going to require sustained effort and clear calls for action 19:27:59 it can't be "what about this?" and wait and see. 19:28:41 http://fedoraproject.org/wiki/Board/Possible_Future_of_Fedora_Governance (old and drafty) 19:29:02 that's all i had. thanks. 19:30:31 * abadger1999 notes that the linked page mostly described what is rather than the discussion of what we'd like it to be. 19:30:56 yeah, couldnt find much better off hand. 19:31:32 nirik: yeah -- now that I read that i remember we got off in the weeds and so what we wrote up in the wiki didn't get into the "exciting" stuff. 19:31:46 yeah 19:31:55 anything else for open floor? 19:32:39 if not, will close in 2 minutes 19:35:04 #endmeeting